Skip to main content

Full text of "Kambuja Desa"

See other formats



co >- 

Q: 00 

_ 68385 




Sir William Meyer Lectures 1942-43 





deliver these lectures an invitation, which I consider to be a high 
distinction and a great privilege. I would also like to offer cordial 
thanks to my esteemed friend Prof. K. A. Nilakanta Sastri M.A., who 
presided over these lectures, and whose company and hospitality 
I enjoyed in ample measure during my stay at Madras, 



MAY 8, 1943. 


1. AymonierLe Cambodge by E. Aymonier, 3 Vols. Paris, 1900-1903. 

2. BCAI.nBulletin de la Commission Archeologique de Flndo-chine. 

3. BEFEO.=iBulletin de 1'Ecole Franchise d 4 Extreme-Orient. 

4. Champa Ancient Indian Colonies in the Far East, Vol. I, Champa by 

Dr. R. C. Majumdar, (Lahore, 1927). 

5. Chatterji=Indian Influence in Cambodia (Calcutta University, 1928)- 

6. Corpus=Inscriptions Sanscrites du Cambodge by M. Barth and A, 

Bergaigne (Paris, 1885). 

7. Et. As.=Etudes Asiatiques (Hanoi, 1925), 

8. Ferrand-Textes=Relations de voyages et Textes Geographiques 

Arabs, Persans et Turks relatifs a I'Extreme Orient by G. Ferrand 
(Paris, 1913-14). 

9. Inscriptions=Inscriptions du Cambodge by G. Coedes (Hanoi, 1937), 

10. Maspero=L'Empire Khmer by G. Maspero (Phnom Penh, 1904). 

11. Suvamadv!pa=:Ancient Indian Colonies in the Far East, Vol. II, 

Suvarnadvipa, (Part I, Political History, Part II, Cultural 
History) by Dr. R. C, Majumdar (Dacca, 1937). 









INDEX .. 163 



I propose to review, in a course of six lectures, the history of 
the Indian colony of Kambuja-desa 1 (modern Cambodia) and some 
aspects of the civilisation that the Hindus, using this term in its 
broadest sense, had introduced in this distant land. I shall try to 
describe how the small isolated Hindu kingdoms in different parts 
of Cambodia were welded into a mighty kingdom that stretched from 
the Bay of Bengal to the sea of China, how the essential spirit of 
Hindu culture was transplanted to this distant corner of Asia, how 
the Hindu religion inspired it to build monuments whose massive 
grandeur still excites the wonder of the world and far surpasses 
anything known so far in India, how art and institutions, created 
on Indian models, grew and developed a unique character, how 
this mighty colonial kingdom flourished for more than a thousand 
years fed by constant streams of civilisation flowing from the 
motherland, and at last met with inevitable decline when this peren- 
nial source itself decayed and ceased to flow. The treatment of 
the subject will necessarily be of a general character, as minute 
discussions of controversial points will be out of place in a public 
lecture. But I shall try to bring together the most reliable data 
available on the subject, and when these series of lectures will be 
published in the form of a book, add notes to explain the different 
view-points and the source and authority of my statements. Two 
considerations have induced me to follow this method. In the first 
place I wish to awaken the general interest in a subject which is 
at present but little known. For although the history of Greater 
India constitutes an important and brilliant chapter of the History 
of India, it has not yet appealed to the general public, and even 
to professed students of Indian history to any considerable extent. 
Secondly, I wish to emphasise the broad features of the history and 
civilisation of Kambuja in order that a solid foundation may be 

1. The term Kambuja-desa, or simply Kambuja has been used to indi- 
cate the ancient Hindu colonial kingdom, in the modern French Protectorate 
of Cambodia, 


laid for further detailed studies on the subject. It may be noted 
that there is at present no text which gives a critical review of the 
history of Kambuja as a whole, in the light of modern researches 
on the subject. 2 It is necessary for a comprehensive study of the 
subject to prepare a skeleton to which flesh and bone may be added 
later. The absence of such a skeleton hampers the efforts to study 
the subject in detail by utilising the abundant data pouring in 
every year from the archaeological researches of the French 
savants. Such a study will be facilitated by the establishment of 
a solid framework which it will be my endeavour to reconstruct in 
course of these lectures. 

Although the history of Hindu colonisation in Cambodia is the 
principal subject of this course of lectures, it is necessary, in order 
to view it in its true perspective, to make a broad survey of the 
state of Indo-China at the moment when the Hindus first came into 
contact with it. This is particularly important, for, as we shall see 
later, the Kambuja empire in its greatest extent embraced nearly 
the whole of this region with the exception of Upper Burma and 
Tonkin. The Hindu culture and colonisation in this vast region 
must be viewed on the background of the land and the peoples in 

2. The following texts deal with the general history of Kambuja. 

1. M. Aymonier Le Cambodge, 3 Vols. The main work is devoted 
to a description of the different localities in Siam and Cambodia 
with notices of the monuments and inscriptions. The concluding 
volume, published in 1904, gives a brief outline of the political 

2. G. Maspero L'Empire Khmer (1904). This is the first systematic 

treatment of the political history of Kambuja. But it is only a 
very brief sketch, the ancient period being comprised in 27 pages. 

3. A. Leclere Histoire du Cambodge A comprehensive history of 

the country from the earliest to modern times. (1914) . 

4. Etienne Aymonier Histoire de Vancien Cambodge (1918) . It is 

a popular treatment of the subject, and was originally published 
in a Paris newspaper. It gives no authority for the statements 

5. Dr. B. R. Chatterji -Indian Cultural Influence in Cambodia (1928). 
The only scholarly work, in English, on the subject. It treats 
the political history and different aspects of culture of Kambuja. 

It will be seen that none of the texts is later than 1928, the year 
memorable for the new theory of P. Stern about the evolution 
of the art of Kambuja, which has practically revolutionised our 
conception of the history and progress of Kambuja culture. A 
number of new inscriptions discovered since 1928 have also pro- 
foundly modified our views about the political history of 


and amid which they flourished. We would, therefore, begin with 
a short account of these two, emphasising particularly those features 
which throw light on, or help the study of, subsequent history. 

The great Indo-Chinese Peninsula covers the whole of the main- 
land of Asia to the east of India and south of China. Shut off by 
the high chains of hills from the continent on the north, it has 
easy means of communication, by sea and land, with both India 
and China. Large and broad at the north it gradually narrows as 
it advances to the south, ending in a long strip of land known 
as the Malay Peninsula. * In addition to this, it covers the region 
now known as Burma, Siam, Laos, Cambodia, Tonkin, Annam and 
Cochin-China. The last four form a French Protectorate, while 
Laos is divided between this Government and Siam. 

The region, distinctly marked off from the rest of Asia, has 
no physical unity like its neighbour, India or China. Situated bet- 
ween two oceanic systems, the Bay of Bengal and the China Sea, 
it has a large coast-line with numerous harbours facilitating contact 
between one another and with the outside world. The interior is, 
however, mostly difficult of access, being intersected by long spurs 
of hills into a number of small plateaus and valleys without easy 
means of inter-communication. The most characteristic physical 
feature of the Peninsula is a series of parallel ranges of hills running 
generally north to south throwing spurs in all directions, with a 
large river, also running north to south, enclosed between each pair 
of hills forming so many water-sheds. On the extreme east and 
west we have two great ranges running along the whole coast line 
and separated from the sea by a narrow strip of plains. One start- 
ing from the south of China constitutes the Annamite Hills, and the 
other from Assam and Manipur passes through Arakan, Tenasserim 
and Malay Peninsula. The other parallel ranges form the water- 
sheds between the rivers Irawaddy (with its branch Chindwin) , the 
Sittang and the Salween in Burma, the Mekong and the Menam 
in the Central region, and the Red River in Tonkin. These rivers, 
carrying silt from the uplands, formed the deltas which constituted 
rich alluvial fertile plains of smaller or bigger size, according to 
varied physical conditions of the country. These deltas on the sea- 
coast formed a striking contrast to the hills and dales of the interior, 
and formed, along with the narrow strips of land in Annam and 
Malay Peninsula between the hill-range and the sea, the strong 
centres of Hindu colonisation in the Peninsula. With the exception 
Df Upper Burma which had direct access from India by land, these 
deltas formed the main strongholds of Hindu culture, and the bases 


from which it radiated, principally along the river valleys, towards 
the interior. It is, therefore, not a mere accident that the most im- 
portant Hindu colonial kingdoms were founded, and the Hindu 
culture and civilisation exercised an abiding influence, mainly in 
these regions, whereas the Hindu influence in the interior was 
comparatively slow, less profound and of shorter duration. 

The people who inhabited the Indo-Chinese Peninsula at the 
time when the Hindus first came into contact with it belonged to 
different races and spoke a number of tongues. Without attempt- 
ing to be too precise and scientific from 'ethnological and linguistic 
points of view, which would require a separate treatment with 
lengthy elaboration of details, beyond the scope of the present re- 
view, I may refer to some of the main classifications which are 
generally agreed to by the scholars. 

First we come across two groups of people, known as Tibeto- 
Burmans and Mon-Khmers, who are generally believed to have 
migrated from India in pre-historic times, and in any case certainly 
show greatest resemblance in physical features and linguistic forms 
with some non-Aryan tribes in India still living in hilly regions 
remote from centres of civilisation. The Tibeto-Burmans consisted 
of a large number of Mongoloid tribes, and those who peopled Upper 
Burma show the greatest resemblance to the Abor and Mishmi 
tribes in Eastern India. 

The designation Mon-Khmer, applied to a group of peoples, 
is derived from the names of its two principal tribes viz., the Mons 
and the Khmers. Their languages belong to the same family as 
those of the Munda and Khasi tribes in India, and the Semang and 
Sakai of the Malay Peninsula. The name Austro-Asiatic is now 
applied to this group of languages, and it is believed that the tribes 
speaking them, at least the Mons and Khmers, originally lived in 
India, and came to Indo-China when they were pressed by invading 
Aryans. The Mons settled in Lower Burma and proceeded thence, 
along the valley of the Menam, to the interior of Siam proper. The 
Khmers peopled Cambodia and moving towards the west met their 
kinsmen, the Mons, in Siam. Their mutual relations would be re- 
ferred to later. 

Two other important groups were the Chams, who lived in 
what is now called Annam, but was known formerly as Champa, 
and the Malays who settled in the Peninsula, now known after 
them. These two belonged to the large group which constitutes 
today the predominant element of the population in Sumatra, Java, 


Bali and other islands of the Indian Archipelago or Indonesia. It 
was recognised long ago that the languages of the Chams and Malays 
belong to the same family as that of Polynesia, and the name 
Malayo-Polynesian was at first applied to this group. Since then, 
however, Melanesian, Polynesian, Micronesian and Indonesian (or 
Malay) languages have all been proved to belong to the same family 
to which the new name Austronesian has been applied. 

So far we are on sure grounds. But there is a theory which, 
though not generally accepted, cannot be omitted in the present 
review, as it concerns the general question of Indian colonisation 
in the Far East, and puts its history in an altogether different light. 
Tpie great German scholar Schmidt who first established the exis- 
tence of the linguistic family called Austro-Asiatic, referred to 
above, has proposed further to connect with it also the Austronesian 
and establish a larger linguistic unity which he calls Austric. He 
also indicates the possibility of an ethnic unity among the peoples 
whose linguistic unity is thus assumed. In other words, Schmidt 
regards the peoples of Indo-China and Indonesia such as the Mons, 
Khmers, Chams and the Malays as belonging to the same race as 
the Munda and allied tribes of Central India and the Khasis of 
North-eastern India. He regards India as the original home of all these 
peoples who, starting from India towards the east, at first spread 
themselves over the whole length of the Indo-Chinese Peninsula and 
then over all the islands of the Pacific Ocean up to its eastern extre- 
mity. This theory, we must remember, has not yet found general 
acceptance among scholars, but we must not lose sight of the possi- 
bility that the Aryanised India, in establishing colonies in the Far 
East, was merely repeating or continuing the work which had been 
inaugurated long long ago by many other peoples inhabiting the 
same land before the advent of the Aryans. 3 

While the coastal regions of Indo-China were thus peopled by 
races whose language bears a strong affinity with that of non- Aryan 
peoples of India, and all or most of whom had probably migrated 
from India in pre-historic ages, the interior of the Indo-Chinese 
Peninsula was dominated, at the beginning of the Christian Era, by 
groups of peoples who belonged to the Thai race. 4 This race is now 
well-known from its most important settlement in Siam, which has 
recently changed its name to Thailand. The Siamese have put a 

3. Tliis topic has been fully treated in my work 'Suvaiiiadvipa', Part I, 
pp. 11 ff . 

4. For an account of the Thais cf. T'oung Pao, 1897, p. 53; 1909, p, 495. 


new interpretation on the word Thai, meaning free, an invention 
designed to emphasise the liberation of their country from the yoke 
of Kambuja in the thirteenth century A.D. The Thailand, however, 
etymologically means, not the "Land of the Free" as the Siamese 
would have us believe, but the land of the Thai tribe. For this 
name Thai has been regularly used, centuries before the indepen- 
dence of Siam, by various other branches of the tribe who had 
numerous settlements in the uplands of Indo-China. As the Thais 
are so far little known in this country, but intimately bound up 
with the history of Kambuja and the expansion of Hindu culture 
in Indo-China, I shall set forth briefly their origin, and later 
discuss their history, so far as it has been ascertained on reliable 

The Thais are a Mongolian tribe and are generally believed to 
be ethnically related to the Chinese. In any case, they, or at least 
a large group or groups of them, lived in southern and south-eastern 
part of the country now known as China. About three centuries 
before the Christian era, or probably somewhat earlier, the Thais 
in large groups migrated to the south and south-west. Two of their 
early settlements were in the regions which we call today Tonkin 
and Yunnan. During the early centuries of the Christian era a 
steady stream of the Thais proceeded towards the west and south- 
west and set up numerous other principalities. The dates and 
gradual stages of their advance cannot be fixed with certainty, 
but by the 8th or 9th century A.D. they had advanced as far as 
the Upper Irawaddy and Salween rivers in the west, and the fron- 
tiers of Siam and Cambodia in the south. 

Thus when the Hindu colonists first came into contact 
with Indo-China, about the beginning of the Christian Era, 
or probably somewhat earlier, they found there peoples of 
diverse races in the primitive state of civilisation, Every- 
thing indicates that the Hindus came by land-routes to upper 
Burma but, for the rest, they mainly followed the sea-route. For 
we find that the Mons, the Khmers, the Malays and the Chams 
were profoundly influenced by these colonists and strong Hinduised 
kingdoms were founded in their lands all along the sea-coast. Thus, 
beginning from the west, we find the Hinduised Mon kingdoms of 
Dhanyavati, Basim, Ramavati, Hamsavatl and Suvarnabhumi (or 
Sudhammavati) on the western and southern coasts of Lower 
Burma corresponding respectively to Arakan, Bassein, Rangoon, 
Pegu and Thaton. Further south, beyond Dvaravati in Siam and 
a number of small kingdoms in the Malay Peninsula were 


the Hinduised Khmer kingdom of Kambuja in Cambodia 
and Cham kingdom of Champa in Southern Annam. All these 
kingdoms bore a strong impress of Hinduism in all aspects 
of their culture and civilisation. Thus excepting Tonkin, which 
came early under the Chinese influence, the rest of the 
coastal regions, and particularly the delta of the Irawaddy, 
Salween, Menam and Mekong rivers were seats of powerful Hindu 

As regards the interior of Indo-China, the Indian colonists 
seem to have settled in large number in Upper Burma, and the 
Hinduised Tibeto-Burmans founded important principalities there. 
This region came early under the influence of Indian culture which 
has still a complete hold of the people. From eleventh century 
onwards the Hinduised Tibeto-Burmans established their political 
authority over the Hinduised Mon kingdoms of the coastal regions 
of Burma, and ultimately the two peoples and cultures were fused 
together, though the Mons, or Talaings, the name by which they 
are better known, still form a distinct element in south-eastern 
parts of the land. The Hinduised Mons, before they were merged 
into the Tibeto-Burmans, spread their cultural influence along the 
coast of Bay of Bengal throughout the Tenasserim region, and down 
the valley of the Menam river and its tributaries up to the very 
heart of modern Siam. There they met with the Hinduised Khmers 
who had established their influence in the lower valley of the 
Menam. The contact between the Hinduised Mons and Khmers. 
and the gradual expansion of the latter towards the north at the 
expense of the Mons and the Thais will be described in course of 
the discussion on Kambuja. 

The Thais, as already noted, peopled nearly the whole 
of the uplands of Indo-China to the east of Burma and 
north of Siam and Cambodia. The region was full of moun- 
tain ranges and dense forests, interspersed with valleys, and the 
rivers which flowed through them were not navigable on account 
of torrents and rapids. The history of Indian colonisation in this 
region and the extent to which it was influenced by the Hindu 
culture and civilisation are but imperfectly known to us, and this 
is partly due to the comparative inaccessibility of a large part of 
this region to modern explorers. But although we are not in a 
position to give a detailed and systematic account of the Hindu 
colonisation in this area, some broad facts may be stated indicating 
its general nature and extent, 


The two most important Thai principalities in Indo-China were 
those in Yunnan and Tonkin. These were the farthest from India 
and nearest to China. It should be remembered that about the 
time when the Thais first settled in these regions the Chinese king- 
dom proper did not extend beyond the Yang-se-kiang river, but its 
rulers tried to extend their political authority over the Thais, 
whom they called barbarians, living in the south and south-west, 
It is unnecessary for our present purpose to describe the long-drawn 
struggle between the two, and it will suffice to say that the Thais 
in Yunnan, though occasionally defeated and subjugated for longer 
or shorter periods, never ceased to defy the authority of the Chinese, 
and ultimately established their independence. By the seventh 
century A.D. they Had freed themselves completely from Chinese 
control, and established a powerful kingdom which played an im- 
portant role in the history of Indo-China for more than six hundred 
years. The kingdom is generally, though not very correctly, desi- 
gnated as Nan-chao, but it is called Videha-rajya, and its capital 
is named Mithila, in the native chronicles. The history of Tonkin 
was more chequered. There the Annamites were subjugated to 
China for a long period, and it was not until the tenth century A.D. 
that they regained their independence and set up a very powerful 
kingdom which comprised not only Tonkin, but also the northern 
part of modern Annam. The Annamites undoubtedly formed a 
branch of the Thais though some are of opinion that they had a 
strong admixture of the Mon blood. 5 The Annamites are desig- 
nated as Yavanas in native chronicles, and adopted Buddhism. 

The effective political authority exercised by the Chinese over 
Tonkin for more than a thousand years resulted in the introduction 
of Chinese culture in Tonkin, and this is the only region in Indo- 
China whose civilisation may be said to have been definitely mould- 
ed by that people. Far different was the case with Yunnan. 
Although ethnically allied to the Chinese and living immediately 
on its border, the Thais in Yunnan seem to have been brought 
under the cultural influence of India, either directly by the Indian 
colonists or indirectly through the Hinduised states in Burma. In 
the absence of epigraphic records and other contemporary evidence 
it is difficult to give a precise or detailed account, but broad general 
indications are not wanting. The great French scholar Pelliot has 
brought together a number of isolated facts and traditions which 

BEFEO. XXI. pp. 260 ff., 274-75, 


seem to prove that the Hindu culture, which has left such a strong 
impress upon Upper Burma, also made its influence felt in Yunnan. 
Although it is generally believed, on the authority of Chavannes, 
that the Thais of Nan-chao were ignorant of writing, Pelliot has 
drawn attention to one or two inscriptions in unknown characters, 
which probably originated in Nan-chao. These characters appear 
to be of Hindu origin. It is characteristic of the Hinduised people 
of Indo-China, that they sought to create a new India by giving 
well-known Indian place-names to their towns and kingdoms. 
According to this practice we find the name Gandhara applied to 
a part of Yunnan. A part of it, as noted above, was also called 
Videha-rajya and its capital was named Mithila, the kingdom being 
sometimes referred to as Mithila-rastra. Local traditions affirm 
that Avalokitesvara came from India and converted the region to 
Buddhism. It is said that when, towards the close of the 8th cen- 
tury A.D., the ruler of this kingdom became enamoured of Chinese 
civilisation, seven religious teachers of India rebuked the king. In 
the first half of the ninth century A.D. a Hindu monk named 
Chandragupta, born in Magadha and therefore designated Magadha, 
led a brilliant career of a thaumaturgist in Yunnan. There was 
in Yunnan the famous Pippala cave, the Bodhi tree, the sacred hill 
Grdhrakuta and many other localities associated with Buddhism, 
A Chinese traveller of the tenth century A.D. refers to a local tradi- 
tion that Sakyamuni obtained the Bodhi near Lake Ta-H in Yunnan. 
The Buddhist influence in Yunnan is still attested by two bells of 
the llth century with inscriptions in Chinese and Sanskrit, The 
king of Nan-chao had the title Maharaja and also another Hindu 
title, which means the king of the east. According to local tradi- 
tion the royal family was descended from the great Asoka. Rasi- 
duddin, writing in the 13th century, not only calls the country 
Gandhara but asserts that its people originated from India and 
China. All these demonstrate that the Thais of Yunnan had imbibed 
Hindu culture and civilisation to a very large extent. 6 

There were many other Thai States to the west and south of 
Yunnan, The Chinese refer to the Brahmana kingdom of Ta-tsin 
to the east of the mountain ranges that border Manipur and Assam, 
and another about 150 miles further east, beyond the Chindwin 
river. Whether these were mainly peopled by Hinduised Thai we 

6. For Nan-chao cf. the detailed account of PelJiot in BEFEO. IV, pp. 
152 ff, where other references are given, 


cannot say. But a group of Thai states, united in a sort of loose 
federation, which occupied the region between the Irawaddy and 
the Salween, was known as Kosambi, The southern part of this is 
now known as the Shan States, the Shan tribe being that branch of 
the Thais which proceeded farthest in the western direction. To 
the east of these were a series of small states extending from the 
frontier of Yunnan to those of Kambuja and Siam. These were, 
from north to South, Alavirastra, Khmerarastra, Suvarnagrama, 
Unmargasila, Yonakarastra, Haripunjaya and many others, whose 
internecine wars, and consequent changes in boundaries and some- 
times also in names, are recorded in the local chronicles, written 
in Pali, of which we possess quite a large number. These Pali 
chronicles give detailed accounts of the ruling dynasties and the 
religious foundations of the different local states. These cannot be 
regarded as historical annals in the sense in which we understand 
the term, but they leave no doubt that the mainspring of the 
civilisation of most of the Thai States lay in India and not in China. 
The evidence of the Pali chronicles is fully corroborated by the 
archaeological finds, for images of the Gupta style and those of 
somewhat later date have been found in these regions. It is a signi- 
ficant fact that these Thais, though ethnically belonging to the same 
race as the Chinese, and living nearer to them, should have been 
brought so profoundly under the influence of Hindu culture and 
civilisation rather than Chinese. 7 

There is no need to feel surprised about the Indian influence 
in these regions of Indo-China. For we have definite evidence 
that as early as the second century B.C. there was regular communi- 
cation, by overland route, between East India and Yunnan. In 
the second century B.C. Chang-Kien, the famous Chinese ambassa- 
dor in Bactria, was surprised to find there Chinese silk and bamboo 
products which, he learnt on enquiry, came from Yunnrn and Sez- 
Chuan across the whole breadth of Northern India right up to 
Afghanistan and Bactria beyond the Hindukush. The two Indian 
Buddhist missionaries who visited China in the first century A.D. 
most probably passed through the upper valley of the Irawaddy 
and Yunnan. There are references also to the regular communi- 
cation between China and Western Asia, via Yunnan, Upper Burma 
and India in the second and third centuries A,D, I-tsing also refers 

7. For an account of the Thai States in central Indo-China cf. Et. As, 
Vol. II, pp. 96 ff. 


to 20 Chinese pilgrims as having gone to India through Yunnan 
and upper Burma. The geographical memoir of Kia Tan, written 
between 785 and 805 A.D., describes two routes leading from Tonkin 
through Yunnan and Burma to India. That this route was well 
frequented in the tenth century A.D. is attested by the fact that 
the 300 religious missionaries sent by the Chinese Emperor to India 
in 964 A.D. in search of sacred texts returned by way of Yunnan. 8 
The large scale raids of the Manipuris in Burma and of the Burmese 
in Manipur prove the use of these routes down to the middle of 
the eighteenth century A.D. Thus although the direct land-route 
from India to hinterland in Indo-China was comparatively little 
known and less used in very recent times, the case was different 
in ancient and medieval periods, and a constant stream of Indian 
emigrants passed through this route to spread Indian culture and 
civilisation in this region. 

Having thus made a broad survey of Indo-China we may now 
proceed to a more detailed discussion of Kambujadesa which forms 
the subject-matter of this course of lectures. This kingdom varied 
in its boundaries at different periods of its history and covered, 
at its greatest extent, the territories which correspond to Siam, 
Cambodia, Laos and Cochin-China and comprised the valleys of 
the Mekong and the Menam. The Kambujadesa proper corres- 
ponds to Cambodia and Cochin-China, comprising the lower valley 
of the Mekong river, south of the island of Khong and the range 
of hills known as Dangrek mountains. 

The valley of the Mekong comprises the whole of modern 
Cambodia with the exception of the three provinces of Kampot 
on the west and Svay Rieng and Thbong Khinum on the east. The 
last two are, however, watered by the two branches of the river 
Vaicos, which are joined to the Mekong across the vast marshy 
plains by innumerable canals, both natural and artificial, and may 
be regarded as its tributaries forming a common Delta in Indo- 

It has been suggested that the name of the river Me-kong is 
derived from Ma-Gahga, the mother Ganges. 9 Whatever we may 

8. For an elaborate discussion of these routes with full reference to 
authorities cf. BEFEO. IV. pp. 131 ff. 

9. The name Mekong or Mekhong is believed to be composed of two 
words, indigenous me meaning chief or mother, and kong, derived from Sans- 
krit Ganga. It would thus be equivalent to mother Ganges. (Leclere-Cam- 
bodgep. 2. f.n.i). 


think of this, there is no doubt that this river played as important 
a role in the history of Kambuja as the Ganges did in the early 
history and civilisation of Northern India, 

The Mekong is to Cambodia what the Nile is to Egypt. It is 
its very life. Its banks supply the habitations of the people and 
its regular annual inundations fertilise the country. The region 
beyond the reach of the flood-water is almost an arid desert. 

From the point, below the rapid of Prah Patang, where the 
Mekong enters Cambodia, it is enlarged, and its bed is nearly 
doubled, by the large marshy depressions running parallel to its 
course, which have been mostly formed by the old beds of the 
river. It covers the country by its ramifications and is joined, 
near Phnom Penh, to the vast lake of Tonle Sap, about 60 miles to 
the north-west, by a wide sheet of water, full of islands. From 
this point of junction the river branches off into two wide streams, 
connected by numerous cross canals forming islands in the inter- 
vening region, till they both fall into the China Sea forming the rich 
delta of Cochin-China. 

When in June the sun rays melt the snow on the Tibetan 
plateau and the waters come rushing down the hill streams, the 
Mekong and its tributaries rapidly rise, cut through their steep 
banks by numerous sluices and overflow the whole region right 
up to the borders of the forest on the ' Highlands '. Then behind 
the steep river banks, marked by fruit trees, gardens and dwelling 
houses, one sees only a vast sheet of water submerging beneath it 
the lakes, the marshes and the plain. It is not till October that 
water recedes and the ground becomes dry enough for cultivation. 

The vast area of ' Lowlands ', annually inundated by the Me- 
kong, forms practically the whole of the inhabited area of Cambodia 
at the present day. In the region north of Phnom Penh, the people 
are settled mostly in groups along the bank of the Mekong and its 
tributaries, or on the borders of the Highlands. In the dry season 
they even temporarily settle in the outlying area for purposes of 
cultivation, but immediately after the harvest is over, they return 
to their homes on the river in time before it is flooded again. 

In the region south of Phnom Penh the habited area is not 
so strictly confined to the riwr-banks. There the people also 
spread here and there, wherever there are high lands fit for culti- 
vation. This region abounds in palm-trees, and viewed from the 

Id temPle 10 kS Hke a ^ Palm - f rest dotted b 


In this region of annual inundation called 'Lowlands', the dis- 
covery of archaeological ruins proves that the modern settlements 
closely correspond to those of old times. Only it appears that in 
the northern part, the modern inhabited area has extended a little 
beyond the old, whereas the case is just the reverse in the south 
(e.g. the province of Ba Phnom). 

It will be shown later that it was precisely in the 'Lowlands', 
and rather to its southern part, that we can trace the earliest habita- 
tion and political and cultural development in Cambodia. Obviously, 
the earliest Hindu colonists chose the region where conditions of 
livelihood were the easiest. The large number of simple brick 
monuments found in this region were probably constructed by the 
first settlers before the sixth century A.D. In that case we must 
hold that the lower valley of the Prek Tonot, the districts of Bati 
and Prei Krebas and part of the district of Treang must have been 
densely peopled in old days, for the remains of the brick temples 
are particularly numerous in this area. Most of the modern temples 
cover the sites of these ancient ones and many of the mounds, covered 
with vegetation, which emerge above the rice fields in the valley 
of the Prek Tenot hide the ruins of these ancient temples. 

The region to the north and west of the 'Lowlands', beyond the 
reach of the flood, may be ^ 'Highlands', although its 

mean height is not very much above the sea-level. It extends up 
to the Dangrek mountains in the north and the hill-ranges of 
Phnom Kravanh and Sang Re to the west. The low grounds of 
this region are full of muddy depressions covered with high thick 
grass, while the higher part is an arid limitless forest. The tropical 
dense forest of tall beautiful trees, with a rich and varied flora 
and bushy soil, is few and far between. The greater part of it is 
covered by a reddish gravel stone, without moss and almost bereft 
of grass, interspersed with vast areas covered with naked sand- 
stone, offering for days a monotonous sight to the eyes of the 
wearied traveller. The rivers in this area are dry for the greater 
part of the year and are full to the brim in the rainy season. 

The whole of this region now lies deserted and uncultivated. 
One may travel for days together without coming across the least 
sign of human beings. Only the deers, buffaloes and wild ele- 
phants roam undisturbed across these arid fields. A few miserable 
hamlets may be seen here and there at the foot of the hills near 
the springs where the descendants of the primitive wild tribes still 
maintain a precarious existence. Otherwise death-like solitude 


reigns supreme where once stately building! stood and a mighty 
empire and civilisation grew. For it is this area which comprises 
in its southern part the whole of the Angkor region where the 
Hinduised Kambuja civilisation reached its high-water mark of 
development and reared magnificent temples and big populous 
cities with strongly fortified walls and gates, grand palaces, tanks, 
parks, and secular structures of all kinds. Human effort and in- 
genuity, after a hard struggle with nature, converted this region 
into a flourishing centre of civilisation, by building roads, canals, 
tanks, bridges and dams. So long as the streams of Hindu colo- 
nists continued to flow and infused vigour and energy into the 
populace, this region continued to flourish, But as soon as they 
were dried up, the people reverted to their old lethargy. Nature 
triumphed, and once more the region relapsed to its old primaeval 
condition. But still the handiwork of man did not altogether perish. 
Gigantic temples and ruins of mighty cities and palaces have sur- 
vived the destructive forces of nature and still tell the tale of a 
bygone age to awestruck travellers in this wild forest. 

So far about the land. We may now pass on to its inhabitants. 
The earliest people who are known to have inhabited the region 
we have just described were the ancestors of the Khmers who still 
form the predominant element of the people of Cambodia. The 
modern name Khmer was used in ancient times also both by the 
people themselves as well as by the foreigners, along with the name 
of Kambuja, of whose origin we shall speak later. The name 
Khmer appears as Kvir and Kmir in the old inscriptions of Champa 
(Annam) and as Comar in the writings of the Arabs, The use of 
this name Comar by the Arabs has been a source of considerable 
confusion, as early writers have identified it sometimes with Cape 
Comorin and sometimes with Kamarupa (Assam). Us identity 
v with Khmer country or Cambodia is now well established. 10 

It is v>ery likely that the country was originally inhabited by 
savage hill tribes whom the Khmers conquered and forced to take 
shelter in hills and jungles. Of this there is no definite evidence. 
But to the north of Cambodia, beyond the Dangrek mountain, lived 
the Laotians after whom the country is still called Laos. They be- 
longed to the Lao race and were mostly savage hill tribes, and still 
retain most of their primitive characteristics. Their settlements 

10. Sometimes the Arab writers themselves seem to confuse Kamarftpa 
with Comar. 


extended up to the outer fringe of the Hindu colonies but they 
generally kept aloof and though influenced by the Hindu colonists 
never attained to any high degree of culture and civilisation. 

As already noted above, the Mons, who inhabited the lower 
valleys of the Irawaddy and the Salween in Burma, extended fur- 
ther south, and formed along with the Laos and Khmers the 
primitive population of modem Siam. Throughout the course of 
history a distinction is noticeable between this heterogeneous 
Mon-Khmer people of Siam ?nd the pure Khmers of Cambodia. 
The Mon-Khmers were intolerant of the political suzerainty of the 
Khmers and always regarded themselves as a rival power. 

The Khmers and Mons thus constituted the principal elements 
of population in the country which constituted the Hindu kingdom 
of Kambuja. At the time the Hindu colonists first settled there 
these people were in an almost semi-savage condition. According 
to the Chinese accounts the people, both men and women, went 
about naked, and decorated themselves with tattoo marks. The 
Chinese expressly state that it was the first Indian ruler who made 
the women wear clothes. 

What attracted the Indians first towards Cambodia it is diffi- 
cult to say. Perhaps it was merely a stage in the course of exten- 
sive colonial enterprises which marked the Indians during the early 
centuries of the Christian Era. The general question of the Indian 
colonisation in the Far East has been discussed by me elsewhere 11 
and need not be repeated here. It will suffice to say that trade, 
missionary spirit and military adventures all contributed towards 
it, and the Indians advanced towards Indo-China both by sea and 
land-routes. Reference has been mc'de above to the establishment 
of Indian colonies not only in Upper Burma but also in the hilly 
regions in the upper valleys of the Irawaddy, the Salween and the 
Mekong rivers, and we have seen how the Indians advanced further 
south along these rivers and established colonies and states in the 
hinter-zone of Indo-China. Whether they advanced in this way 
as far as Cambodia proper along the banks of either the Menam 
or the Mekong rivers, we cannot definitely say. But we may re- 
call in this connection a passage in Hiuen Tsang's Travels the full 
significance of which is not often realised. After finishing the des- 
cription of Samatata which corresponds 1'oughly to Southern and 

11 . Champa Introduction, 


Eastern Bengal the pilgrim remarks: "Going north-east from this 
to the borders of the ocean we come to the kingdom of Shi-li-cha-ta- 
lo (Srikshetra) ", and he names in succession five other kingdoms 
which were not visited by him, but of which he gained information 
at Samatata. All these kingdoms have not been satisfactorily iden- 
tified, but two of them I-shang-na-pu-lo or Isanapura and Mo-ha~ 
chan-po or Mahachampa undoubtedly correspond to Cambodia and 
Annam. 12 It is a legitimate inference from the statement of Hiuen 
Tsang, and particularly the context in which it is made, that there 
was a regular intercourse by land between E. India and these remote 
regions before the seventh century A.D. The reference to Isanapura, 
is specially significant, for the Kambuja king Isanavarman, 
after whom it was named, ruled in the second and third de- 
cades of the seventh century A.D., and was thus almost a contem- 
porary of Hiuen Tsang. That his name was known in E. India 
before 638 A.D. when the Chinese pilgrim visited it, certainly 
shows that Indians had a fairly regular communication with Cam- 
bodia. The intercourse between Cambodia and Burma is also re- 
ferred to in the early annals. Recent political events have reawak- 
ened our interest in the Burma-India land route, but, as already 
noted above, it seems to have been fairly well-known and regularly 
used more than a thousand years ago. The Burmese annals prove 
that throughout the medieval period Burma had regular intercourse 
through land with E. India on the one hand and Siam, Cambodia 
and Annam on the other. There is thus no inherent difficulty in 
presuming a connection between E. India and Cambodia by over- 
land route. Indeed one scholar seems to have been so much con- 
vinced of the facility of this route that he seriously suggested that 
the Kambuja dynasty which temporarily occupied Northern, and 
Western Bengal in the tenth century A.D. came from Cambodia. 13 

Whatever we ftiay think of a possible land-route to Cambodia, 
there can be no doubt that there was communication by sea from 
very early times. The Periplus of the Erythraean Sea proves that 
at least as early as the first century A.D., ships from Indian ports 
regularly sailed to Malay Peninsula, and there are indications that 
the sea-route to China via Straits of Malacca was also in use, 

12. Beal, II, 199-200; Walters, H, 187-189. For the identifications cf. 
JRAS. 1929 pp. 1,447. Hindusthan Review, July, 1924. 

JHQ. II, p. 250; IV, p. 169; Ind. Ant. LVIII, p. 57; LV; p. 113. 

13. Chatterji, p. 279, 


Ptolemy's accounts indicate further progress of this maritime inter- 
course as he refers to various Indian place-names in Indo-China, 
Malay Peninsula and the islands of the Indian archipelago. One 
of the Chinese chronicles, the History of the Liang dynasty, com- 
piled during the first half of the seventh century A.D., explicitly 
refers to Indian ambassadors coming by the Southern Sea to China 
during the period 147-167 A.D. 14 As the vessels in those days kept 
close to the coast as far as possible, Southern Cambodia must have 
furnished one or more important halting stations in the distant 
voyage between India and China. We must, therefore, presume 
lhat the Indian mariners possessed a knowledge of Cambodia at 
the beginning of the Christian era, if not earlier still. As a matter 
of fact the Chinese chronicles make definite references to maritime 
intercourse between India and Cambodia in the third century A.D. 
As we shall see later, according to local traditions preserved in a 
Chinese chronicle of the third century A.D. the first Hindu kingdom 
was founded in Southern Cambodia in the first century A.D. This 
date cannot in any case be very wide of the mark. 

The beginnings of Indian colonies in Cambodia, like those in 
other parts of Indo-China, are lost in oblivion, but are echoed in 
local legends and traditions. These legends and traditions cannot, 
of course, be regarded as true chronicles of events, but they possess 
historical importance inasmuch as they have preserved the popular 
beliefs about the foundation of Hindu civilisation, and indicate in 
a general way the process of Hindu colonisation of these lands. 

The two most important kingdoms in Cambodia in the earliest 
period were Fu-nan and Kambuja. Both of these had their own 
local legends about the beginning of Hindu colonisation. The legend 
current in Fu-nan as recorded by K'ang Tai in the middle of the 
third century A.D. runs as follows: 

"The sovereign of Fu-nan was originally a female called 
Lieu-ye. There was a person called Huen-chen of Mo-fu. He was 
a staunch devotee of a Brahmanical god who was pleased with his 
piety. He dreamt that the god gave him a divine bow and asked 
him to take to sea in a trading vessel. In the morning he went to 
the temple of the god and found a bow. Then he embarked on a 
trading vessel and the god changed the course of wind in such a 
that he came to Fu-nan. Lieu-ye came in a boat to plunder 

14. BEFEO. HI, pp, 271-72. 


the vessel Huen-chen raised his bow and shot an arrow which 
pierced through the queen's boat from one side to the other. The 
queen was overtaken by fear and submitted to him. Thereupon 
Huen-chen ruled over the country". 15 

The same story is repeated in later Chinese texts, 16 in some 
cases with additional details, such as the marriage between Huen- 
chen and Lieu-ye. The names of the king and queen are variously 
written as Huen-huei or Huen-tien and Ye-lieu. Huen-tien and 
Lieu-ye may be accepted as the correct forms. Huen-tien and the 
other variant forms represent the Indian name Kaundinya. Lieu-ye 
probably means "Leaf of Willow". 

In an inscription in the neighbouring kingdom of Champa dated 
657 A.D. we find an echo of the same story. Referring to the 
foundation of Bhavapura, the capital of Kambuja, it says: 17 

"It was there that Kaundinya, the foremost among Brahmanas, 
planted the spear which he had obtained from Drona's son Asvat- 
thama the best of Brahmanas. 

There was a daughter of the king of serpents, called Soma, who 
founded a family in this world. Having attained, through love, to 
a radically different element, she lived in the habitations of man. 

She was taken as wife by the excellent Brahmana Kaundinya 
for the sake of (accomplishing) certain work. Verily, incompre- 
hensible is the way of God in providing conditions leading to future 

(King Bhavavarman) who, being born in that pure unbroken 
line of kings, is, even to-day, the pride of his subjects by his 
unblamable (conduct)". 

We may compare this legend with the following account pre- 
served in the Cambodian Annals about the origin of the kingdom 
of Cambodia 18 : 

"Adityavamsa, king of Indraprastha, was displeased with one 
of his sons and banished him from the state. He came to the country 

15. Et As, II, pp. 244 ff. Pelliot discusses in this connection the location 
of Mo-fu, the original home of Huen-chen, but is unable to come to any 
definite conclusion. 

16. BEFEO. III. pp. 254, 256, 265. 

17. Champa, Book IH, p. 23. 

18. Champa, p. XVHI, 


of Kok Thlok and made himself master of it by defeating the native 
king. One evening he was walking on a sand bank when suddenly 
the tide arose and obliged him to pass the night there. A Nagl of 
marvellous beauty came to play on the sand and the king, over- 
powered by her charm, agreed tc marry her. Then the Nagaraja, 
the father of the betrothed girl, extended the dominions of his 
would be son-in-law by drinking the water which covered the coun- 
try, built a capital for him and changed the name of the kingdom 
into that of Kamboja". 

A somewhat different version is preserved about the origin of 
Kambujadesa, in later Annals. 

In the dim past, so runs the story, Cambodia was a desert of 
sand and rocks. One day Kambu Svayarnbhuva, the king of Arya- 
desa, found himself in this dreary landscape. The death of his wife 
Hera, whom the great god Siva himself gave to him, made him dis- 
consolate and he left his country "in. order to die in the wildest 
desert 1 ' he could find. Having reached Cambodia he entered into 
a cave. To his horror Kambu found himself in Ihe midst of a large 
number of huge, many-headed snakes, whose piercing eyes were 
turned towards him. Kambu, however, boldly unsheathed his sword 
and advanced towards the biggest snake. To the utter amazement 
of Kambu the snake spoke in a human voice and asked his where- 
abouts. On hearing Kambu's story the serpent said: "Your name 
is unknown to me, stranger, but you spoke of Siva, and Siva is my 
king, as I am the king of the Nagas, the great snakes. You seem 
to be courageous too; therefore abide with us in this land you have 
chosen and end your grief." Kambu remained and came to like 
the Nagas who could take human shape. Several years later he 
married the Naga king's daughter. The king of the Nagas possess- 
ed magic power and turned the arid land into a beautiful country 
like that of Aryadesa. Kambu ruled over the IShd and the kingdom 
came to be called after him 'Kambuja'. This mythical legend is 
briefly referred to in the Bakesei Cankrom Ins. dated 947 A.D., 
where the Kambuja kings are said to have been descended from the 
great sage Kambu Svayarnbhuva, to whom Hara gave as wife Mera 
the most glorious of Apsaras. 

It is interesting to note that both the mythical traditions noted 
above were current among the Pallavas who ruled in South India 
in the early centuries of the Christian Era. Thus some records 
describe Skandasisya, the progenitor of the Pallavas, as the son of 
Asvatthama (son of Drona) by a Naga woman. Other records 


refer to VIrakurcha, the predecessor of Skandasisya, as having 
married a Nagi and obtained from her the insignia of royalty. 
Manimekhalai and three other Tamil texts also mention the marri- 
age of a Chola king and a Nagi and their son as the Pallava king 
of Kaiichl The basic factor in all these traditions viz., the origin 
of a royal dynasty from the marriage of an Indian prince with a 
Naga woman, is thus similar to the traditions current in Fu-nan, 
and what is more striking, the mythical Asvatthama is associated, 
though in different role, in both cases. The tradition may, however, 
be carried still further back. For Herodotus records a similar 
story about the progenitor of the Scythians. Heracles, we are told, 
met in Scythia a strange creature, the upper part of whose body 
was that of a woman but the lower part that of a serpent. She 
bore three sons to Heracles, who left instructions before his depar- 
ture that whichever of these three could bend the bow loft by him, 
should be made the king of the country. According to this test 
Scythe became the ruler and his descendants were called after him 
Scythians. Apart from general resemblance the episode of the 
bow, contained in it, offers the common element with one version 
of the Fu-nan tradition preserved in Chinese texts. Whether the 
common tradition about the origin of the royal dynasty proves any 
close connection between the people of Fu-nan and the Pallavas 
on the one hand, and between the Pallavas and the Scythians on 
the other, is a speculative problem, which it is unnecessary to 
discuss for our present purpose. We must, however, note that there 
is also a Pallava parallel to the second tradition of Kambuja which 
derives the royal dynasty from the sage, Kambu Svyambhuva, to 
whom the God Siva gave as wife the most glorious Apspra named 
Mera. The Pallava story as recorded in a Sanskrit inscription found 
at Amaravati runs as follows. "By the favour of Siva, Drona had 
a glorious son named Asvatthama who became an ascetic, and lived 
in a forest. One day the Apsara Madani came to his hermitage, 
and both became enamoured of each other. The Apsara bore him 
a son named Pallava, who became the originator of the dynasty 
known after him." Here also we find this common basic theme, 
viz., the origin of the royal dynasty by the union between a sage 
and an Apsara, though the role played by Siva is slightly different. 
The mention of Asvatthama is also interesting inasmuch as he 
figures in the other tradition. 19 

19. BEFEO. XI, pp. 391-93; XXIV, pp. 501 ff. 


Whatever we might think of these traditions as a connecting 
link between the Indian colonists in Cambodia and a known region 
or a ruling dynasty in India, they have undoubtedly some historical 
value, for we cannot fail to note in them an allegorical representa- 
tion of the conquest of the land of primitive wild tribes (Nagas) 
by the colonists from India (Aryadesa) who introduced the ele- 
ments of higher civilisation among the primitive aborigines. The 
introduction of names like Kambu and Siva is no doubt due to the 
dominance of Saivism. 

Whether the same band of colonists spread over the whole of 
Cambodia or its different parts were colonised by different groups 
of immigrants at different times cannot be definitely decided. Nor 
is it quite certain whether all the immigrants came direct from 
India or some of them at any rate came from other Indian colonial 
kingdoms. The probability is that a number of small states were 
founded in different parts of Cambodia by groups of people coming 
at different times both from India and Indian colonies in the neigh- 
bourhood. The names of some of these states which acknowledged 
the suzerainty of Fu-nan are preserved in Chinese chronicles. 

The History of the Liang Dynasty (502-556 A.D.) gives us the 
following account of Tueri-siun. 

"In the southern frontier of Fu-nan, at a distance of more than 
3,000 li lies the kingdom of Tuen-siun on a rugged sea-shore. It 
is about 1000 li in extent. The capital city is about 10 li from the 
sea. There are five kings, all of whom are vassals of Fu-nan. The 
merchants from India and Parthia came in large numbers to carry 
on trade and commerce, the reason being that Tuen-siun forms 
a curve projecting into the sea for more than a thousand li. The 
sea 20 is vast and without limit and cannot be crossed directly. 21 
Hence the market of Tuen-siun forms a meeting ground between 
the east and the west, frequented every day by more than ten 

20. The Chinese name Chang -hai refers to the Sea of China extending 
from Hai-nan to the Malay Peninsula, It includes the Gulf of Tonkin 
(BEFEO. Ill, p. 263, f.n. 2) 

21. As Pelliot observes, the Chinese text is somewhat obscure. Both 
Pelliot and Schlegel are of opinion that the Chinese vessels did not directly 
cross the sea from Annamese coast to the Malay Peninsula. They consequently 
followed the coast line and probably the goods were transhipped across the 
isthmus of Kra, thereby saving a lot of time. There is, however, no definite 
evidence in support of this view (BEFEO. Ill, p. 263 f.n. 3 cf., Suvdrnadwpa, 
I. pp. 85-6). 


thousand men. Rare objects, precious merchandises, in short 
everything is found there. Moreover there is a tree, resembling 
pomegranate-tree, the juice of whose flowers is collected in a jar 
and, after a few days, is transformed into wine". 22 

An Indian, named Che by the Chinese, who lived in the fifth 
century A.D. 23 and visited these parts gives the following account 
of Tuen-siun. 

"Tuen-siun is a vassal state of Fu-nan. The king is called 
Kuen-Luen. It contains five hundred Hu (probably of mercantile 
caste) families of India, two hundred Fo-tu (probably Buddhists), 24 
and more than thousand Brahmans of India. The people of Tuen- 
siun follow their religion and give them their daughters in marriage, 
as most of these Brahmans settle in the country and do not go 
away. Day and night they read sacred scriptures and make offer- 
ings of white vases, perfumes and flowers to the gods. When they 
fall ill they take a vow to be devoured by the birds. They are led 
outside the town to the accompaniment of music and dance, and 
they are left to be devoured by the birds. The bones are then 
burnt and put in a jar which is thrown into the sea. If the birds 
do not devour them, they are placed in a basket. The ''cremation 
by fire" consists in throwing the body into fire. The ashes are col- 
lected in a vase which is buried, and to which sacrifices are made 
for an unlimited period". 25 (It then refers to preparation of wine 
as in the previous extract). 

The account of Tuen-siun is very illuminating as it reives a 
vivid image of an Indian colony in a foreign land, and shows the 
process by which Indian colonies grew and exerted their influence 
over the indigenous population. It is the usual story of trade 
followed by a missionary propaganda, both Brahmanical and Bud- 
dhist, of gradual settlement of Indians in the country, and ultimate 
fusion with the people by intermarriage with the native population. 

Like Tuen-siun Fu-nan itself was an important market town 
where met the traders from India and China, Evidently the same 
process, as described above, did also operate here in making it a 

22. BEFEO. Ill, p. 263. 

23. BEFEO. IE, p. 277. 

24. Fo-tu means Buddha as well as Stupa. The figure is two, possibly 
a mistake for two hundred (Ibid, p, 279 f ,n. 5) . 

25. Ibid, p. 279. 


flourishing Indian colony. Its supremacy was perhaps due to the 
fact that Indian colonists seized the political power at an early stage. 

References are not wanting in Chinese texts to communication 
between Fu-nan and India, The story of Kia-sing-li, who came 
from India to Fu-nan in course of trade, will be mentioned later. 
Reference has already been made to the fact that during the period 
147-167 A.D. several Indian embassies went to China through the 
Southern seas. As the vessels in those days kept close to the coast, 
Fu-nan must have been an important halting station in this voyage. 
The new History of the Tang dynasty (618-906 A.D.) refers to ex- 
change of diamonds, sandals and other goods between India on the 
one hand and Fu-nan and Kiao-che (Tonkin) on the other. 26 An- 
other Chinese text says that a big vessel of Fu-nan corning from 
western India had for sale a mirror of blue po-li (spluttika) which 
had a diameter of 16 ft, 5 inches and weighed more than forty 
pounds. 27 

We may thus easily visualise the process by which Fu-nan 
and other Indian colonies were founded in Cambodia. In course 
of time Fu-nan grew more powerful than others and established its 
sway over the whole of Cambodia and even far beyond its frontier. 
Fortunately we possess a somewhat detailed account of its rise and 
development, and this will form the subject of my next lecture. 

26. Ibid, p. 275. 

27. Ibid, p. 283. 



The earliest historical account of the Khmers is bound up with 
the kingdom of Fu-nan, which is frequently referred to in the 
Chinese texts from the third to the seventh century A.D. Strangely 
enough, all traditions of the kingdom, and even its very name, dis- 
appeared after the seventh century A.D. without leaving any trace. 
The location of this kingdom became, therefore, a matter of con- 
siderable difficulty, and various opinions about it were expressed 
by different scholars. To M. Paul Pelliot belongs the credit of 
finally settling this question and also of collecting together the his- 
torical information about this kingdom scattered in different Chinese 
texts. The following account of the kingdom of Fu-nan is mainly 
based on the materials collected by him on the subject. 1 

The name of the kingdom, Fu-nan, is written in different ways 
by the Chinese, and I-tsing calls it Pa-nan. M. Aymonier regards 
it as a pure Chinese word meaning 'the protected south/ but Pelliot 
infers from the different forms of the name, that it is merely a 
Chinese transcription of an original name. Unfortunately, this 
original name cannot be restored with any amount of certainty. 
The views of Schlegel and Parker that the original name was P'o- 
nam or Phnom ( penh) are rejected by Pelliot. 2 G. Coedes 
derives the name from Ba Phnom, a region round the hill of that 
name in South Cambodia, 3 and this seems to be the most plausible 

The kingdom of Fu-nan corresponded roughly to Cambodia 
proper and a part of Cochin-China, and comprised the lower valley 
of the Mekong. The capital of this kingdom, according to some 
Chinese texts, was 500 li from the sea. If, as it seems very pro- 
bable, this distance was measured by the route along the Mekong 
river, the capital must have been situated between Chaudoc and 

1. Cf. "Le Fou-Naii" by P. Pelliot in BEFEO. Vol. Ill, pp. 248-303. In the 
following foot-notes of this chapter, the pages, unless otherwise indicated, 
refer to this article. 

2. P. 288. 

3. BEFEO. Vol. XXVIII, pp. 129-30, 


Phnom Penh, a region which was once the centre of Khmer civili- 
sation. Pelliot, following Aymonier, infers from an inscription 
found in the province of Battambang that the earliest capital of the 
historical kingdom of Kambuja which replaced Fu-nan was most 
probably Vyadhapura. He identifies it with modern Angkorbaurei, 
and thinks it very likely that it was also the capital of Fu-nan. 4 
Coedes, however, rejects this view. He locates Vyadhapura at the 
foot of the peak at Ba Phnom and regards it as the capital of Fu- 
nan, though not of Kambujadesa. 5 

According to the Chinese account the primitive people of Fu- 
nan were semi-savages. They went about naked and decorated 
themselves with tattoo marks. 6 Their queen Lieu-ye was, how- 
ever, defeated by Huen-tien, a follower of the Brahmanical religion, 
who introduced the elements of civilised life, In particular he 
made the women wear clothes. 7 

This Huen-tien was most probably a Hindu colonist who came 
direct from India, though the possibility is not altogether excluded 
that he might have been a Hinduised colonist from some part of 
Malay Peninsula or Malay Archipelago. 8 From the accounts of 
subsequent events 9 his arrival cannot be placed later than the first 
century A.D. No particulars of Huen-tien's reign are known to 
us, 10 but his son is said to have been given an appanage of seven 
towns 11 . The existence or creation of these vassal states was not 
without danger to the kingdom. However, one of the successors 
of Huen-tien, named Huen P'an-huang, sowed the seeds of dis- 
union among the seven towns and thus succeeded in bringing them 
under his control. Then he appointed his sons and grandsons as 

4. P. 290. 

5. BEFEO. Vol. XXVin, pp. 127-131, 

6. P. 265. 

7. P. 256. See ante p. 17. 

8. Cf. Et. As. II, pp. 245-46, where Pelliot discusses the location of 
Mo-fu, the original home of Huen-chen (pp, 247-69), but is unable to come 
to any definite conclusion. 

9. These are referred to below. 

10. Except, of course, his civilising mission referred to above. 

11. Cf. p. 265. The passage is somewhat obscure, though this seems to 
be the proper meaning. It is evident from what follows that the governors 
of the seven towns grew to be too powerful for the central authority and 
it was only by creating dissensions among them that they were once more 
brought under control, 


governors of only a single town. They were called 'small kings' 12 
Huen P'an-huang, as the first part of the name indicates, was un- 
doubtedly a descendant of Huen-tien, 13 and ruled during the second 
half of the second century A.D. 

Huen Pan-huang died at the advanced age of ninety and was 
succeeded by his second son P'an-p'an. He left the cares of govern- 
ment to his great general Fan-man, or Fan-che-man. When the 
king died after a reign of three years Fan-che-man was elected 
king by the people 14 (c. 200 A.D.) . 

Fan-che-man was an able ruler and laid the foundation of 
the greatness of Fu-nan. He constructed a powerful navy and 
conquered the neighbouring states to a distance of five or six 
thousand li which henceforth became vassals of Fu-nan. Although 
the Chinese names of these vassal states cannot all be satisfac- 
torily identified, we may hold in a general way that nearly the 
whole of Siam and parts of Laos and Malay Peninsula acknowledg- 
ed the authority of Fu-nan which thus became the first Hindu 
colonial empire in Indo-China. Fan-che-man assumed the 
title 'Great king of Fu-nan' and was about to lead a campaign 
against Kin-lin (Suvarnabhumi or Suvarnadvipa) when he 
fell ill and died. During his illness he had sent his 
eldest son Fan-Kin-cheng to take charge of the army, but 
the general Fan Chan, son of the elder sister of Fan-che-man, taking 
advantage of the absence of Fan-Kin-cheng, declared himself king 
and put Fan-Kin-cheng to death (c. 225 A.D.). 15 

The reign of Fan Chan is of special importance as we know 
definitely that he sent embassies to both China and India. Accord- 
ing to San Kuo Che (which deals with the history of the period from 
220 to 280 A.D. and was written towards the end of the third cen- 

12. P. 265. 

13. This also clearly follows from the statement in the History of the 
Southern Tsi Dynasty that the sons and grandsons of Huen-tien ruled the 
country up to the death of P'an-huang (pp, 256-7). 

14. P. 265. 

15. C/. p. 266-7 and the footnotes. For Kin-lin cf. p. 266 f.n. (5). It was 
about two thousand li from Fu-nan. The people were Buddhist and there 
were several thousand sramanas in the country. The Chinese Kin-lin means 
'frontier of gold* but it is sometimes identified with Kin-chen which means 
island of gold. The Chinese name may, therefore, be regarded as equivalent 
to Suvarnabhumi or Suvarnadvipa. On the identification of this cf. my book 
1 Suvarnadvipa, Part I pp, 37 ff. 


tury A.D. by Chen Chen) Fan Chan, king of Fu-nan, sent an 
embassy to China in 243 A.D., offering as presents a few musieians 
and some products of the country. 16 This is one of the earliest 
references to official relationship between China and Fu-nan. 17 In- 
cidentally it gives us the first fixed point in the chronology of the 
kings of Fu-nan, on the basis of which it has been possible to assign 
approximate dates to the previous kings. 

A Chinese text of the third century A,D. tells us that Kia- 
sing-li, an inhabitant of T'an-Yang, in the western part of India, 
made voyages for purposes of trade and ultimately reached Fu-nan 
during the reign of Fan Chan. He gave the king a graphic des- 
cription of the laws, manners, customs, and the immense wealth 
of India. Being asked the distance, he said it was 30,000 li, and a 
return voyage from Fu-nan would take three to four years, 18 Fan 
Chan's curiosity was perhaps aroused by these stories. In any 
ca^e he sent one of his relations named Su-Wu as an ambassador 
to India. Su-Wu embarked at Teu-ki-li, probably the famous 
port of Takkola, and reached the mouth of the great river of 
India (Ganges) after about a year. Having proceeded up the river 
for 7,000 li, he met the king of India. The latter cordially wel- 
comed Su-Wu and arranged for his visit to the different parts of 
the kingdom. He senl two envoys to accompany Su-Wu to the 
king of Fu-nan with a present of four horses of Yu-che country, 
and these came to Fu-nan four years after Su-Wu had left the 
country. 19 

These four years, however, witnessed great political changes. 
King Fan Chan was no longer on the throne of Fu-nan, He was 
assassinated by Fan Chang, a younger son of Fan-che-man. Fan 
Chang was a baby at the time of his father's death, but when he 
was twenty years old, he collected a few brave persons and killed 
Fan Chan in order to avenge the murder of his elder brother. It 
is not definitely known whether Fan Chang ascended the throne, 
but even if he did so, his reign must have been short. He was 

16. P. 303. 

17. Pelliot says on p. 503 that this is the earliest embassy, but on p. 2$3 
he refers to a passage from Wu-li to the effect that Fu-nan sent an embassy 
to China in 225 A.D. (or, according to another version of the same passage 
of Wu-li, quoted by a later writer, during the period 229-31 A.D.) 

18. P. 277. 

19. P. 271. 


assassinated by the general Fan Siun who succeeded him as king 
of Fu-nan. 20 

It was during the reign of Fan Siun, probably some time bet* 
ween 245 and 250 A.D., that the Chinese ambassadors K'ang T'ai 
and Chu Ying visited Fu-nan. it was evidently in recognition of 
the embassy sent by him to China. The Chinese ambassadors met 
in Fu-nan Chen-song, one of the envoys sent by the king of India. 

This Chinese embassy is of more than passing importance to 
Fu-nan. K'ang T'ai and Chu Ying wrote two books on Fu-nan 
which gave to the Chinese the first authentic accoum of the king- 
dom and supplied the main source of information to later writers 
on the subject. 21 In particular K'ang T'ai's work ib frequently re- 
ferred to by later authors. He may be compared to Magasthenes, 
and it is of tragic interest to note that like the Indica of the latter, 
K'ang T'ai's work is lost and only fragments of it are preserved in 
quotations by later authors. It may be added that K'ang T'ai also 
recorded a brief account of India as reported by Chen-song. 

The only point of importance in this account of India is that 
it gives us more detailed information about the particular kingdom 
in India visited by the envoy of Fu-nan. According to the state- 
ment of Chen-song, as reported by K'ang-T'ai, 'the title of his king 
was Meu-lun and to the right and left of his kingdoms, there were 
six great kingdoms, viz., those of Kia-wei (Kapilavastu) , Che-Wei 
(Sravasti) etc. This leaves no doubt that the great river of India 
through which the envoy of Fu-nan proceeded inland for 7,000 li 
was the Ganges, and that the kingdom he visited was situated 
somewhere in U.P. M. Sylvain Levi has proposed the identification 
of Meu-lun with the Murundas. This dynasty is referred to in 
the Puranas as having ruled for 350 years, but there is no indica- 
tion about the locality of their kingdom. A Jairia work refers to 
Pataliputra as the residence of a Murunda-raja, but it is not clear 
whether it is a personal (cf. Sisunaga king Miuida) or a tribal 
name. The Murundas are also referred to in the Allahabad Pillar 
inscription of Samudragupta, but it is extremely improbable that 
they could have ruled anywhere in the upper Ganges valley at the 
time of that great emperor. S. Levi's hypothesis that the envoy 
of Fu-nan visited the kingdom of the Murunclas near about Pa^ali- 

20. P. 267. 

21. P. 275. 


putra, is at best a plausible, though very doubtful one, and all that 
we can safely assert is that Fu-nan's ambassador visited a kingdom 
on the upper Ganges valley. 

As regards K'ang Tai's account of Fu-nan, the only point that 
need be referred to at present is his observation that though the 
country is beautiful, it is strange that the men went about naked. 
King Fan Siun, however, stopped this indecent habit. 22 

Fan Siun had a long reign and sent several embassies to China 
in the years 268, 285, 286 and 287 A.D. 23 An interesting sidelight 
is thrown on the political status of Fu-nan about this time by the 
memorandum prepared by T'ao Huang, the governor of Tonkin, 
when the emperor expressed a desire to reduce the military ex- 
penditure. T'ao Huang prayed that the garrison of Tonkin which 
formerly consisted of 7,000, and now only 2420, men should not 
be reduced any further. As a reason for this he pointed out the 
danger of constant incursions of Fan Hiong, the ruler of Champa. 
Further, he said, that the Chams and the people of the adjoining 
country Fu-nan are allies, -and the two support each other. Their 
tribes are numerous and they don't submit to China. 24 

The next reference to Fu-nan in Chinese history is in connec- 
tion with an embassy sent in A.D. 357 by a Hindu named Chan- 
tan. 25 According to the Chinese texts this Hindu took the title 
of the King of Fu-nan. This indicates a period oi' political troubles 
with several claimants for the throne. The name of the Hindu 
may be restored as Chandana or Chandra. 26 He sent a petition 
and presented some tamed elephants to the Chinese king. The 
latter, however, issued an order to the effect that as the mainten- 
ance of these animals entails considerable expenditure they should 
not be sent as presents. 27 According to other texts the emperor 
considered these strange animals as sources of evil to the people 
and ordered them to be returned* 28 

22. P, 268. 

23. P, 252. 

24. P. 255. 

25. Pp. 252-3, 255, 269. 

26. S. Levi restores the name as Cmasthana equivalent to Devaputra, and 
regards it as a title of the king of India who in his opinion sent this embassy. 
(Melanges Charles de Harlez, pp 176 ff). Pelliot rejects this view (p. 252 
f.n. 4). 

27. P. 269. 

28. P. 255* 


Towards the end of the fourth or the beginning of the fifth 
century A.D. the throne of Fu-nan was occupied by Kiao-chen-ju 
or Kaundinya. The History of the Liang Dynasty has preserved 
the following story about him: "Kaundinya was a Brahman and 
an inhabitant of India. One day he heard a supernatural voice 
asking him to go and reign in Fu-nan. He reached P'an-p'an to 
the south of Fu-nan. The people of Fu-nan cordially welcomed 
him and elected him king. He introduced Indian laws, manners 
and customs/' 29 This story perhaps preserves an echo of a fresh 
stream of colonists coming direct from India, as a result of which 
the country was thoroughly Hinduised. 

Next we hear of Ch'e-li-t'o-pa~mo, a successor of Kauncjinya 
sending embassies, with presents, to the Imperial Court in 434, 435 
and 438 A.D. 30 

The History of the First Song dynasty which gives us the above 
information also tells us that in the year 431 or 432 A.D. the king 
of Champa, intending to overthrow Tonkin, asked for military aid 
from the king of Fu-nan, but the latter refused the request. 31 The 
king of Fu-nan was most probably Ch'e-li-t'o-pa-mo. 

The Chinese texts tell us a great deal more about another suc- 
cessor of Kaundinya. 32 

Towards the close of the Song period (420-478 A.D.) king Chb- 
ye~pa~mo (Jayavarman) ruled in Fu-nan. His family name was 
Kaundinya. He sent some merchants to Canton for purposes of 
trade. On their return journey the Indian monk Na-kia-sien (Na- 
gasena) joined them for coming back to his country. But a storm 
forced them to land in Champa whose people plundered all their 
goods. Nagasena, however, reached Fu-nan. 

In A.D. 484 Jayavarman sent Nagasena to the imperial court 
with a long petition, the full text of which is given in the Chinese 
chronicles. After the usual compliments and expressions of good 
will it refers to the disastrous voyage of Nagasena and the mer- 
chants from Canton in course of which they were robbed by the 

29. P. 269. 

30. The History of the first Song Dynasty which refers to the embassies 
in detail gives the name of the king as Ch'e-li-pa-mo. (p. 255). The History 
of the Liang Dynasty gives the name as Che-li-to-pa-mo (p. 269), 

31. P. 255, 

32. Pp. 257-261. 


king of Champa. It then refers to the glowing account of the laws, 
religion and the government of China given by Nagasena which 
induced Jayavarman to send his humble presents and ask for the 
good wishes of the emperor. 

The petition then narrates in detail how a rebellious subject of 
Fu-nan, named Kieu-ch'eu-lo fled to Champa, organised a rebellion 
there and made himself master of Champa. He was there indulg- 
ing in all sorts of violence and injustice, and what was worse, 
adopted an attitude of open hostility against the king of Fu-nan, 
his original master. As Fu~nan and Champa had a common boun- 
dary, Jayavarman was naturally anxious to get rid of him and asked 
the emperor to send a force against Champa, which he compla- 
cently described as originally a vassal state of China. He offered 
to help the imperial troops in their task of subjugating Champa, 
and agreed to recognise, as King of Champa, any other person nomi- 
nated by the emperor. Even if the emperor were unwilling to 
send a powerful army to chastise the king of Champa, Jayavarman 
requested him to send a small force to help him in punishing the 
wicked king. In order to strengthen his case he sent rich presents 
including a golden model of the throne of Naga-raja, an elephant 
of white sandal, two ivory stupas, two pieces of cotton, two vases 
of precious transparent stones, and a betel-nut plate made of shell. 

Nagasena proceeded to the imperial capital and gave an account 
of the manners and customs of Fu-nan, the most interesting point 
in which is a reference to the dominant cult of Mahesvara, the god 
who lives on the Mot an hill. He also presented a poem, which is 
somewhat abstruse but evidently eulogises the god Mahesvara, 
Buddha and the emperor. 

The emperor praised the god Mahesvara and -condemned the 
wicked usurper of the throne of Champa. But then he added: "It 
is only by the culture and virtue that I attract the distant people, 
but I do not like to have recourse to arms. However, according 
to the established convention of the government, I am referring the 
request of the king of Fu-nan for military -assistance against Champa 
to a tribunal/' The decision of the tribunal is not on record, but 
there is nothing to show that any military assistance was given. 
The emperor, however, presented a large quantity of silk of various 
colours to the king of Fu-nan. 

In 503 A.D. Jayavarman again sent an embassy to the imperial 
court with presents including an image of Buddha, made of coral. 
On this occasion the following imperial edict was issued: "The 


king of Fu-nan Kaundinya Jayavarman lives on the border of the 
ocean. For generations he and his forefathers have ruled over 
that distant country of the south and their sincerity (of respectful 
feelings to the emperor) is manifested by frequent embassies and 
presents. It is proper to show some favour in return and to bestow 
a grand title upon him. Hence (I confer) the title "The General 
of the pacified south, the king of Fu-nan." 33 

Jayavarman sent two more embassies to the imperial court, one 
in 511 and the other in 514 A.D. 34 There is no doubt that through- 
out his reign a very cordial and intimate relation subsisted between 
the two countries. This is further proved by the fact that two 
Buddhist monks of Fu-nan settled in China 35 whose works are still 
preserved in the Tripitaka. One of them, Sanghapala or Sangha- 
varman (460-524 A.D.), knew several languages, and spent sixteen 
years (506-522) in translating, at the command of the emperor Wu, 
various canonical texts in five different places. One of these was 
called Fu-nan-Kuan or Bureau of Fu-nari. 

The second monk was named Mandra or Mandrasena. He 
arrived at the imperial capital in A.D. 503, and was commanded 
by the emperor Wu to collaborate with Sanghapala in the transla- 
tion of sacred scriptures. 

Jayavarman died in A.D. 514. His elder son, Rudravarman, 
born of a concubine, succeeded him after having killed the younger 
son born of his legitimate wife. An inscription found at Neak Ta 
Dambang Dek in the province of Treang 36 in southern Cambodia 
refers to the foundation of a hermitage (drama) with a tank and a 
dwelling house (dlaya) by queen Kulaprabhavati, the principal 
spouse of a king called Jayavarman. In view of the palaeography 
and find-spot of the Ins. Coedes, who edited it, has identified this 
king with Jayavarman of Fu-nan. The alphabet of this inscription 
bears a close resemblance with that of the Thap Musi Ins. 37 of 
Gunavarman. who is described as the young son of a king of the 
family of Kaundinya. Coedes suggests on the joint evidence of 
these two inscriptions that the young Gunavarman was the son of 
Jayavarman and Kulaprabhavati, and his legitimate succession to 

33. P. 269. 

34. P. 270, 

35. Pp. 284-5. 

36. Edited by Coedes, JGIS, IV, pp. 117 ft 

37. BEFEO. XXXI, pp. 1 ff. 



the throne was prevented by Rudravarman. This is a very reason- 
able hypothesis, but cannot be regarded as an established fact until 
further evidence is available. 

Rudravarman is also referred to in an epigraphic record, 38 A 
Buddhist inscription, sadly mutilated, belongs to his reign, and 
eulogises his royal qualities. It refers to the appointment of an 
official by his father Jayavarman, but does not give us any histori- 
cal information. 

Rudravarman sent no less than six embassies to China in 517, 
519, 520, 530, 535 and 539 A,D. The envoy sent in 517 was an In- 
dian named Tang-pa o-lao (Dharmapaia?). The presents sent in 
519 included an image of Buddha made of Indian sandalwood, and 
pearls or precious stones of India. In 539 he sent a living rhinoceros 
and offered to the emperor a hair of Buddha 12 ft. long which was 
in his country. The emperor sent a monk to fetch the precious 
relic. 38a 

Rudravarman is the last king of Fu-nan referred to by name 
in the Chinese texts. Nothing is known of this kingdom during 
the next three quarters of a century. But we learn from the 
Chinese chronicles that Fu-nan was conquered by Citrasena king 
of Chen-la, whose son Isanasena sent an embassy to the Chinese 
court in 616-7 A.D. 39 It is obvious, however, that the conquest 
by Chitrasena did not mean an end of the kingdom of Fu-nan. For 
we learn from the Chinese texts that T'6~mu, the capital of Fu-nan, 
was suddenly seized by Chen-la, and the king of Fu-nan removed 
himself to a town called Na-fu-na, fuiiher to the south. 40 Accord- 
ing to Polliot the Chinese name may stand for Navanagara, and 
this city was probably situated somewhere near Kampot. 41 Further, 
the Chinese texts refer to two embassies of Fu-nan in the first half 
of the seventh century A.D. 42 The last reference to Fu-nan occurs 
in the account of I-tsing (671-695 A,D.) in the following words: 

38. Ibid. 

38a. Pp. 270-1. 

39. P. 272. It is not expressly stated that Isanasena sent the embassy, but 
that is the obvious inference as he is referred to as the reigning king. 

40. P. 274, 

41. P. 295. 

42. One between 618 and 626 A.D., and the other between 627 and 649 AD, 
(p. 274). 


"Leaving Champa and going towards the south-west the country 
of Pa-nan is reached. Formerly this was called Fu-nan. In anci- 
ent times it was the country of the naked men. The people wor- 
shipped many Devas. Then the law of Buddha prospered and 
expanded. But at the present time a wicked king has completely 
destroyed it and there are no more monks/' 43 

It may be surmised from the above that some vestige of the 
ancient kingdom of Fu-nan survived till the end of the seventh 
century A.D. 

The name Chen-la is used by the Chinese to rcier to the king- 
dom of Kambuja. The inscriptions of Cambodia give u-; a detailed 
and connected account of the kingdom from the seventh century 
A.D. Even the two kings of Chen-la, referred to above, are known 
from these inscriptions. There can be hardly any doubt that Kam- 
buja was originally a vassal stale of Fu-nan that grew powerful 
enough in the seventh century A.D. to assert its supremacy and 
destroy the suzerain power. 44 Henceforth Kambuja takes the place 
of Fu-nan and continues a glorious existence for nearly seven 
hundred years. But before proceeding to deal with the history of 
this powerful kingdom wo should make a broad review of the king- 
dom of Fu-nan, the first Hindu kingdom in Cambodia, if not in 
Indo-China, and trace the development of its culture and civilisation 
on the basis of the data supplied by the Chinese chronicles. 

The early history of Fu-nan is a repetition of that of almost 
every ancient Hindu colony in the Far East. Originally a country 
of savages or semi-barbarians, it imbibes the element of civilisation 
from a Hindu or Hinduised chief who establishes his authority 
either by conquest or by more peaceful methods. Gradually it 
comes more and more into direct contact with India and Hindu 
culture and civilisation become the dominant feature. 

In the case of Fu-nan we can distinctly trace two broad stages 
of Indianisation, one in the first and another in the fourth century 
A.D., and in both cases under the influence of its rulers, whose 
names are supposed to represent the same Indian namo, Kaundinyn. 
The earlier, Huen-tien, is said to have followed the Brahmanical 
cult, but there is no definite information of his original home. 
Naturally one might be tempted to regard him as coming direct 

43. Takakusu-I-tsing, p. 10. 

44, This will be more fully dealt with in the next Lecture, 


from India. But the story of the Indian merchant Kia-siang-li who 
visited the court of king Fan Chan seems to militate against this 
view. For, if we believe in the details of the story it would appear 
that the king had never heard of India before. But such stories 
are not to be taken too literally, and perhaps the sentiments of 
the Chinese writer, to whom India was a terra incognita, rather 
than those of the king and people of Fu-nan, have been reflected 
in them. For the Chinese texts tell us that during the time of 
the emperor Huan of the later Han dynasty (147-167 A.D.) Indian 
embassies came to China by the southern sea and they must have 
passed by Fu-nan. According to the same texts, however, this 
political intercourse completely ceased in the third century A.D. 45 
This might account for the ignorance of India in the Far East in 
the third century A.D., but as the story of Hucn-tien goes back 
to the first century A.D. there is no inherent improbability in the 
very natural assumption that he came from India. We have, there- 
fore, good reason to believe that the first Hindu colonists from India 
settled in Fu-nan not later than the first century A.D. 

There is, however, no doubt that Kaundinya of the fourth cen- 
tury A.D. came direct from India, as this is explicitly stated in the 
Chinese Texts. 

The first reference to the people of Fu-nan occurs in a poem 
composed in the third century A.D. by Tso-Sseu. He says that the 
people of Fu-nan 46 are clever and should not be confused with the 

The earliest general account of Fu-nan is given in the History 
of the T$in Dynasty which covers the period from 265 to 419 A.D. 
and was composed by Fang Hiuan-ling (578-648 A.D.). It runs 
as follows; 

"The kingdom of Fu-nan is more than 3,000 li to the west of 
Lin-yi (Champa) in a great bay of the ocean. The country is 
three thousand li in extent. There are many walled towns, palaces, 
and houses. The people are black and ugly. They have curly hair 
and go about naked and bare footed. Their nature is simple and 
they are not at all given to theft or robbery. They apply them- 
selves to agriculture. They sow one year and gather harvest 
during next three years. Moreover they love to engrave and 

45, Pp. 271, 272, 
46, P. 281, 


chisfel their ornaments. They mostly take their food on silver 
plates. The taxes are paid in gold, silver, pearls and perfumes. 
They have many books and there are libraries and archives. In 
writing they use an alphabet derived from India. 47 Their funeral 
and marriage ceremonies are like those of Champa' '. 48 

The following account is preserved in the History of the 
Southern Ts'i covering the period from 479 to 501 and composed 
at the beginning of the sixth century A.D. 

"The people of Fu-nan are crafty and malicious. They forcibly 
carry away and reduce to slavery the inhabitants of the neighbour- 
ing towns that do not submit to their authority. Their articles of 
trade are gold, silver, and silk. The men of noble families use 
Sarong made of brocade. The females cover their body in a dress 
that passes over their head. The poor people cover their bodies 
with a piece of cloth. The inhabitants of Fu-nan use golden rings 
and bracelets and silver vessels. They cut woods to make their 
houses. The king lives in a storeyed pavilion. The houses are 
enclosed by a wooden palisade. The houses are sometimes covered 
by bamboo-leaves, 8 or 9 ft. long. They also live in houses raised 
above the ground. They construct boats 80 to 90 ft. long and 6 
or 7 ft. wide, the front and back of which are shaped like the head 
and tail of a fish. When the king goes out he rides on an elephant. 
The women also ride on elephants. They arrange cock-fight and 
pig-fight for their amusements. They have no prison. In case of 
dispute they throw a golden ring or egg in boiling water and the 
disputants have to draw them out or they have to walk seven steps 
carrying red-hot iron chain in their hands. The hands of the 
guilty are completely burnt, but the innocent do not suffer any 
injury. Sometimes the disputants are thrown into water. The 
guilty sinks but the innocent does not. The country produces 
sugar-cane, pomegranate, orange and much areca-nut. The birds 
and the mamalions are the same as in China. The character of the 
people is good and they do not like war". 49 

47. The actual words are "their characters resemble those of Hu". But 
as Pelliot points out (p. 254, f.n. 2), that although the Hu, properly speaking, 
mean the people of Central Asia, all alphabets related to those of India are 
included in the Hu alphabet. That the Indian alphabet was used in Fu-nan 
is definitely proved by the discovery of the Sanskrit inscriptions referred to 
in f.n. 36-38 above. 

48. P. 254. 

49. Pp. 261-2. 


The beginning and the end of this account give us somewhat 
contradictory ideas about the general disposition of the people. 
Either they refer to different types of people living in the same 
country, or put together observations made by different persons, 
perhaps at different times. 

The History of the Liang dynasty (502-556 A,D.) repeats much 
of the above information but also adds something new. The pro- 
ducts of the country are said to be gold, silver, copper, tin, aloe, 
ivory, peacock, kingfisher and parrots of five colours. It pdds a 
few details about the trial by ordeal referred to above. We are 
told that crocodiles are kept in the ditches outside the walls, and 
ferocious animals are placed inside an enclosure outside the city 
gates. The persons accused of criminal acts are thrown in the 
ditches and the enclosure, and kept for three days. If they are 
not devoured, they are considered innocent. In all cases the sus- 
pected persons are made to fast and practise abstinence for three 
days before being subjected to the ordeal. 

The big crocodiles are said to have been more than 20 ft. long. 
They resembled the alligators, had four legs and their mouths, 
six or seven feet long, had on each side teeth pointed like swords. 
They lived ordinarily on fish, but would devour a buck or a man 
if they could get it. 50 

It gives the following account of the manners and customs of 
the people: 

"They do not sink wells in their houses, several dozens of 
families having a common pond whence they draw their water. They 
worship the spirits of heaven and make their images in bronze. 
Some of these have two and four faces with respectively four and 
eight hands. Each hand holds something, a child, a bird, an animal, 
sun or moon. 

"The king and his concubines who live in the palace ride on 
elephants when going out or coming in. When the king sits, he 
squats on a side, with the right knee raised high and the left knee 
touching the ground. In front of him is spread a piece of cotton 
cloth, and vases of gold and burning incense are placed on it. 

"In case of mourning the custom is to shave the beard and the 
hair. The dead are disposed of in four ways, the dead body being 

50. Pp. 263, 268. 


cither burnt, thrown in a river, buried in the ground, or left in 
the field for being devoured by birds. 

"The people are greedy by nature. The boys and girls mix 
promiscuously and they observe neither ceremony nor decency." 51 

The New History of the Tang dynasty (618-906) refers to a 
kind of diamond produced in the country. In appearance it is like 
csuartz. It is found in abundance on rocks deep under water and 
the people dive into water to collect them. A jade can be scratched 
by it, but it is broken if struck by a corner of a battering ram. 52 

Another Chinese text, Wai Kuo Clnian, refers to a new type 
of trial by ordeal. If anything is stolen in a house, a small quantity 
of rice is taken in -a temple and the god is requested to find out 
the thief. The rice is placed near the feet of the god. Next day 
the rice is taken and the servants of the house are asked to eat it. 
The innocent swallows it without any difficulty, but blood comes 
out of the mouth of the guilty. 

The same text tells us that the people of Fu-nan are noble and 
wealthy, and decorate thfcir houses with sculpture. They are gene- 
rous and charitable and have many birds and animals. The king 
is fond of hunting. They aril ride on elephants and spend days and 
months in hunting excursions. 52 

The Tong-tien, a text of the nature of an Encyclopaedia, com- 
posed at the end of the eighth century, contains the following 

"In the time of the Suei (581-618 A.D.) the king had the 
family name Ku-long. There were many others in the kingdom 
having the same family name. The old people, when interrogated, 
says that the K'uen-luen (Malays) have no family name. (The 
Ku-long) is a corruption of K'uen-luen." 54 

The accounts culled above from the Chinese texts leave no 
doubt that the people of Fu-nan had imbibed Hindu culture and 
civilisation to a very large extent. The dominant religion was 
Saivism as we learn from the account of Nagasena, though Buddhism 

51. P. 269, 

52. P. 274. 

53. P. 280. 

54. P. 283. 


was not unknown. The images of gods with two or four faces and 
four or eight hands undoubtedly refer to Brahmanical gods. Some 
of their manners and customs, such as putting on clothes, riding 
on elephants, funeral ceremonies, trial by ordeal, the luxurious 
mode of living as indicated by their decorated houses, costly orna- 
ments and utensils, the royal pomp and grandeur, 55 the peculiar 
shape of their boats, and the title Fan (equivalent to Varman) 
borne by the kings all indicate an Indian influence. Their alpha* 
bets were also derived from India. 

The traces of primitive barbarism were no doubt still to be 
found among a large section of people. This perhaps explains 
some of the contradictory statements in Chinese accounts. For 
while some say that they had marriage ceremonies like those preva- 
lent in Champa, others refer to the promiscuous relation between 
boys and girls and want of a sense of decorum and decency among 
the people. Again, as we have already noted above, the same 
account which praises the good character and pacific disposition of 
the people, also refers to their ferocious and violent character. All 
these seem to indicate that although the people generally gave up 
their primitive nomadic habits and barbaric customs and took to 
agriculture, they had not all imbibed the culture and civilisation 
introduced by the Hindu colonists to the same degree. This is 
only natural, and the same phenomena are observed in other Hindu 
colonies e.g. Champa and Java. The main credit of the Hindu 
colonists lies in the great contrast between the primitive barbarism 
of the people whose men and women went about naked, and the 
highly developed culture and civilisation that flourished in the same 
soil after the settlement of the Hindus. 

But the best evidence of the extent to which Indian culture 
was imbibed by the people is furnished by the three Sanskrit ins- 
criptions of Fu-nan. 56 The first begins with an invocation to God 
Visnu. The second records a donation by prince Gunavarman to 
the image of God Vi&nu called Cakratirthasvami which was con- 
secrated by the Brahmanas versed in the Vedas, Upavedas and 
Vedangas and sages versed in the Sruti. The third records a 
donation to some Buddhist establishment and refers to Buddha, 

55. The History of the Liang Dynasty tells us that king Fan Siun con- 
structed pavilions and belvederes and held three or four audiences a day 
(pp. 267-8). 

56, For these inscriptions cf. f.n. 36-38 above, 


Dharma and Sahgha. The Bhagavatas are mentioned in the second 
and the Aryasamghikas in the third. Thus while the Chinese 
evidence refers to Saivism and Buddhism, the first two inscriptions 
definitely prove that Vaisnavism was also introduced in the coun- 
try. The second also demonstrates that the Indian philosophical 
ideas and religious beliefs were familiar to the people of Fu-nan. 
Referring to the temple of Visnu, erected by Gunavarman, it says: 

"Tad-bhakto=dhivased vised-api ca vd tustdntardtma. jano \ 
Mu/cto duskrta-karmmanas~sa paraman-gacchet padam 
vaisnavam \ \ 

"A devotee of Vinu, who lives in, or even once enters into, 
this place consecrated to the god Cakratirthasvami, with a conten- 
ted heart, will be freed from the effects of his evil deeds, and go to 
the abode of Visnu". The cult of Bhakti and the theory of Karma 
are thus clearly referred to. We are further told that the sages 
versed in the Sruti have given the name Cakratirthasvami to the 
God. The name Cakratirtha is mentioned in several Puranas, and 
a knowledge of the Puranic literature can be easily presumed from 
this as well as the mention of Ksiroda-samudra and amrta in the 
stanza of invocation. There is nothing to be surprised at this. For 
the record refers explicitly to Brahmanas versed in Veda, Upaveda 
and Vedanga, and the first inscription mentions the houses of Brah- 
manas in a town called Kurumbanagara. 

The third inscription refers to two kings of Fu-nan named 
Jayavarman and his son Rudravarman. The royal eulogy is com- 
posed in the right Indian style, and the following may be cited as 
an example: 

"Ekasthdn akhildn narddhipagundn udyacchate veksitum \ 
DJidtrd nirmita eka eva sa bhuvi Sn Rudravarmma | | 

Sarvam saccaritam krtam nrpatind ten-dti-dharm-drthiiw, \ 
Lokanugraha-sadhaimm prati na ca ksatravratam khanditam \ \ 

"God created Rudravarman in this world in order to collect in 
one place all the royal virtues. For the sake of dharma he, the 
king, performed all virtuous acts. But he did not forsake the duty 
of a Ksatriya which contributes to the welfare of men". King Jaya- 
varman, we are told, appointed the son of the leading Brahmana 
(dvija-ndyaka) as his treasurer (dlutnanam-adhyaksa) . This Brah- 
mana official adopted Buddhism and became an Upasaka. The 


invocation to Buddha and the general tenor of the inscription show 
no trace of Mahayana influence whatsoever. 

The records show that the caste-system, at least in its general 
form, was introduced in Fu-nan, although the Brahmajias were 
by no means confined to duties of a sacerdotal character. The ins- 
criptions, which are written in Sanskrit verse and in South Indian 
alphabets, show that the Sanskrit language and literature were 
already highly developed on the soil of Fu-nan. Although the 
inscriptions, particularly the last two, are fragmentary, enough 
remains to show that the religion and mythology of India had been 
carried to Cambodia and the essential elements of Hindu culture 
were thoroughly established in the colony of Fu-nan long before 
the sixth century A.D. How thoi-ougHy Indian mythology was 
cultivated in Fu-nan may be illustrated by the description o Queen 
Kulaprabliavati, in the first inscription, as "Saci is that of Sakra, 
Svaha is that of Fire, Rudrani that of Hara, and Sri of Sripati." 

Fu-nan may be regarded as one of the important centres from 
which Indian culture and civilisation radiated on all sides. The 
conquests of Fan-che-man and the establishment of an empire of 
Fu-nan must have facilitated the natural process by which the 
culture was spreading in all directions. The empire of Fu-nan must 
have extended over Siam, Malay Peninsula and a part of Burma, 
in the time of Fan-che-man. Wan-chen, a writer of the third 
century A.D., says that the vassal states of Fu-nan wore all 
governed by Mandarins, and that the great officers of the right and 
the left of the king were called Kuen-luen. This title was borne 
by the king and a large number of families in Fu-nan. These must 
have formed the aristocracy and the ruling class, and they were 
the main instruments of the spread of Hindu culture and civilisation 
all over the empire. 

The same conclusion is borne out by the study of the art of 
Kambuja. The monuments and sculptures of Kambuja fall readily 
into two broad divisions, the primitive and the classic. The latter 
is associated with Angkor and dates from about the 10th century. 
The primitive art begins from the age of Fu-nan and is continued 
by the early rulers of Kambuja which took its place in the 7th 
century A.D. As most of the monuments of the early period were 
made of perishable materials like wood or brick, there are not 
enough remains to reconstruct the art of Fu-nan. But it is now 
generally recognised that the primitive art of Kambuja originated 
during this period, and Parmentier has tried to piece together the 
data furnished by the primitive art in order to give some idea of 


the lost art of Fu-nan. 57 The teni A \* of brick, consisted of 

a square or rectangular cella, with plain walls surmounted by a 
roof which consisted of a number of gradually receding stages. 
This is a characteristic of the Gupta art, and although extant speci- 
mens are wanting, we get representation of it in sculptures of the 
Gupta period. 58 The affinity with the Gupta art is more evident 
in th * sculptures, and the scholars are in general agreement that 
the primitive art of Kambuja is derived from India and is closely 
allied to the Gupta art. 59 In recent yearc a number of sculptures 
have been discovered in Siam and Cambodia who^e style is sur- 
prisingly akin to that of the Gupta art. Robert Dalet has described 
an image of Buddha found at Tuol Prah That in the district of 
Kompon Spur. 60 The image is broken and was found interred 
under the earth. The photograph shows this image to be strikingly 
similar in style to the Sarnath images of Buddha with which we 
are so familiar. This image and others of similar style found at 
Vat Prah Nirpan 61 leave no doubt that the primitive art of Kambuja 
was a direct product o f the Indian school. Indeed Groslter 62 has 
even advanced the theory that the original Indian colonists 
brought with them aitists and craftsmen from India, and they were 
entrusted with the task of building temples and images of gods. 
In short, whatever difference of views there might be about the 
origin of the classical art of Kambuja dating from 9th and 10 cen- 
turies A.D., the scholars are agreed in their view that the art of 
Fu-nan was purely Indian, and through Fu-nan this Indian art of 
the Gupta age spread over a wide territory in Indo-China along 
with other phases of Indian culture. 

57. BEFEO. XXXII. pp. 183 ff. 

58. For fullei discussion of this point cf. Suvarnadvipa, Part II, pp. 347 ff. 
History oj Bengal (Dacca University) Edited by R. C. Majumdar, pp. 493 ff. 

59. cf. G Groslier La Sculpture Khmere Ancienne, Ch. IX. Rene Grousset 
India (Tr, by C. A. Philips) pp. 306 ff. CoedesKecueil des Ins. du Siam 
pt II. p. 4; Ars Asiatica, XII, p. 23. 

60. BEFEO. XXXV, pp, 156 ff. 

61. Ibid, p. 157. 

62. Op. Cit. 



The kingdom of Karri bujadesa, the mythical legend of whose 
foundation has already been mentioned above, rose irto importance 
in the sixth century A.D. and overthrew Fu-nan. Lince that time 
this kigndom prospered and continued its glcrious existence for 
well-nigh seven hundred years, till the inevitable decline set 
in, and it was gradually reduced to a petty protectorate of the 
French in the 19th century A.D. 

This continuity and the glory and splendour of Kambuja have 
cast into shade we might almost say oblivion the earlier king- 
dom of Fu-nan, with the result that the history of the Khmers now 
practically begins with Kambuja and their early traditions are 
bound up with it. 

Kambuja, as noted above, was one of the many states which 
acknowledged the suzerainty of Fu-nan. Its early history is 
obscure but we may glean a few important facts from a study of 
the inscriptions and Chinese accounts. 

The Baksei Camkron Ins. (No. 89) dated 869 &aka (^947 
A.D.), which gives the traditional account of the descent of the 
Kambuja kings from the great sage Kambu Svayambhuva, refers 
to Srutavarman as the root (mula) of the rulers of Kambu who 
delivered the country from bondage (lit. chains of tribute). An- 
other inscription, found at Ta Prohm (No. 177) and dated 1108 S' 
(^1186 A.D.), mentions king Srutavarman and his son king Sres- 
thavarman. The latter is described as the sun in the sky which is 
the family of Kambu, and is said to have been the progenitor of the 
royal line (va$udhddhara-vamsa~yoni) . He was born on the moun- 
tain Jayadityapura and was the supreme king of Sresthapura. 

We may, therefore, regard 6rutavarman and his son {Srestha- 
varman as the earliest historical kings of Kambuja. Fortunately, 
the reference to Sresthapura enables us to locate the kingdom over 
which they ruled. An inscription (No. 170) found at Vat Phu, 
near Bassac in Laos, records that the locality was included in the 
district (visaya) of Sresthapura. It would thus appear that origi- 
nally the kingdom of Kambu occupied this region to the north-east 


of Cambodia. 1 It is perhaps possible to locate definitely the capital 
of this kingdom. According to the History of the Sui dynasty the 
capital of Chen-la was close to a hill called Leng-kia-po-po on the 
summit of which there was a temple. To the east of the town 
lived a spirit named Po-to-li to which human sacrifice was offered. 
The Chinese name of the hill can easily be restored as Lingapar- 
vata, which according to Ins. No. 33 (V. 4) was the ancient name 
of the Vat Phu Hill. The Chinese name of the spirit may be derived 
from the first two letters of Bhadresvara, the presiding deity of Vat 
Phu. An inscription (No. 170) also refers to the country of 
Bhadresvaraspada as included in the Visaya of Sre^thapura. There 
can be hardly any doubt, therefore, that Srethapura, the capital of 
the earliest kingdom of Kambuja, was in the immediate neighbour- 
hood of Vat Phu. 2 

It may be legitimately inferred from what has been said above 
that Kambuja was civilised by the Hindu immigrants who had 
possibly come at different times and set up different kingdoms, all 
of which at first acknowledged the suzerainty of Fu-nan. Sruta- 
varman, the first historical king of Kambuja known to us, evidently 
united some of these petty states into a powerful kingdom and freed 
it from the yoke of Fu-nan. Srethavarman succeeded his father 
Srutavarman and ruled as a powerful king at Sre^thapura near 
Vat Phu. 

The statement in the Baksei Camkron Ins. (No. 89) that the 
kings of this family delivered Kambu from bondage requires further 
elucidation. There can be hardly any doubt that the bondage 
refers to the suzerainty of Fu-nan. But who delivered Kambu 
from the yoke of Fu-nan? He may be either Srutavarman or 
Sre$iavarman or one of his successors of whom nothing is known. 
But the independence seems to have been achieved before a new 
dynasty founded by Bhavavarman succeeded them in Kambuja. 

The Ta Prohm Ins., referred to above, mentions this kin 
Bhavavarman as lord of Bhavapura, and he is expressly stated to 
be the founder of a line of kings. This is corroborated by V. 5 of 
the Ang Chumnik Ins. (No, 27) which says that he got his king- 
dom by his own prowess (sva-sakty-<ikrdnta~rdjya) . His father 
Vlravarman is referred to in two inscriptions (Nos. 9 and 11) ? but 
there is nothing to indicate that the latter was a king. Thus evi- 

1. BJEFEO. XVIII. No. 9. pp. 1-2. 

2, BJSFEO. XXVIIL p. 124. 


dence, both positive and negative, confirms the statement in the 
Ta-Prohm. Ins., that Bhavavarman was the founder of a new line 
of kings. 

Important light is thrown on the history of the royal 
line founded by Bhavavarman by the succession of kings referred 
to in the Ang-Chumnik Ins. (No. 27). This inscription mentions a 
family of physicians and the names of kings served by them. It 
begins with two brothers Brahmadatta and Brahmasirhha who were 
physicians of king Rudravarman. Their nephews (sister's sons) 
viz., Dharmadeva and Simhadeva, were physicians respectively of 
kings Bhavavarman and Mahendtavarman. Simhavira, son of 
Dharmadeva, was a minister of king Isanavarman, and lastly 
Siihhadatta, son of Simhavira, was the physician of king Jaya- 
varman and governor of Adhyapura. 

The inscription thus gives a list of succession of kings as follows: 






This list is partly confirmed by an inscription of Prakasadhar- 
man king of Champa dated 658 A.D. which refers to Bhavavarman, 
king of Kambuja, and his two immediate successors, his brother 
Mahendravarman and the latter's son Isanavarman. 3 The relation- 
ship of the first two kings in the series cannot be ascertained, but 
Bhavavarman was certainly not the son of Rudravarman, for his 
father's name, as noted above, was Viravarman. 

Mahendravarman, according to the Ins. No. 11, was the son of 
Sri- Viravarman and the younger brother of IJhavavarman. The 
same inscription tells us that his original name was Citrasena, but 
he assumed the name Mahendravarman at the time of his coronation. 

The name Citrasena is known to us from the Chinese texts. 
The History of the Sui Dynasty (589-618) gives the following 
account of Chen-la. 4 

3. Champa, Bk> III. pp. 18-19, 23-24. 

4. The account given in the History of the Sui dynasty is translated 
by M, Abel Remusat in Nouveaux Melanges Asiatique, Vol. I, pp. 77 ff. 

(Paris 1829). It refers to the period 589-618 A.D. All subsequent references 
to Chinese Chronicles, unless otherwise stated, are from this work. 


"It is situated to the south-west of Lin-yi (Champa). It was 
originally a vassal state of Fu-nan. The family name of the king 
was Ksatriya. His personal name was Citrasena. His ancestors 
had gradually increased the power of the kingdom. Citrasena 
made himself master of Fu-nan. He died. His son Isanasena 
succeeded him. He inhabited the town of Isana". The same text 
adds that the first embassy from Chen-la came in 616 or 617 A.D. 
and it was obviously sent by Isanasena. 

Another Chinese text Nan-che tells us that the king Ksatriya 
Isana, at the beginning of the period Cheng-kuan (627-649 A.D.), 
conquered Fu-nan and took possession of the kingdom. 5 

The discrepancy in the Chinese texts, one attributing the con- 
quest of Fu-nan to Citrasena and the other to his son Isana, may 
perhaps be best reconciled by supposing that the conquest of Fu-nan 
was a gradual process. The struggle with Fu-nan must have com- 
menced even before the reign of Citrasena, for the History of the 
Sui Dynasty states that the power of the kingdom of Chen-la was 
increased by his ancestors. The only reasonable and natural in- 
ference from -all these is that Bhavavarman began to aggrandise 
the kingdom of Kambuja at the -expense of Fu-nan, and his two 
successors, viz., his brother Citrasena or Mahendravarman and 
the latter's son Isanasena or Isanavarman, completed the conquest 
of the country. It is probable that Citrasena conquered the nor- 
thern part of the state, but the capital of Fu-nan was occupied, 
and that kingdom was finally conquered, by the last named king. 6 

It is not difficult now to identify Rudravarman, the first king 
and predecessor of Bhavavarman in the succession list furnished 
by the Ang Chumnik Inscription. As we have seen above, Rudra- 
varman is the name of tjje last king of Fu-nan mentioned in the 
Chinese chronicles, and his last known date is 539 A.D. As Bhava- 
varman was removed by one generation from his predecessor 
Rudravarman, and as Isanavarman, the son of his brother, was on 
the throne before 617 A.D., there can be hardly any doubt that 
king Rudravarman, served by the two physicians Brahmadatta 
and Brahmasimha, as mentioned in the Ang Chumnik Ins., was the 
king of Fu-nan of that name. 

5. BEFEO. Ill, p. 275. 

6. This has been treated with fuller details in Lecture II, Cf. also 


Thus on the basis of the epigraphic and Chinese data we may 
reconstruct the history of the dynasty of Bhavavarman somewhat 
as follows: 

Some time about the middle of the sixth century A.D. Bhava- 
varman acquired the throne of Kambuja and considerably increased 
its power and extent. His brother Mahendravarman, who succeeded 
him, led many military expeditions against Rudravarman, the king 
of Fu-nan. He succeeded in conquering nearly the whole of the 
kingdom and probably even seized its capital. Rudravarman, or his 
successor who was then on the throne of Fu-nan, fled to the south 
and his dynasty continued to rule over a petty state in the extreme 
south of Cambodia with a new capital city. But the struggle bet- 
ween the two powers continued during the reign of Isanavarman, 
the son and successor of Mahendravarman. Isanavarman finally 
extinguished the kingdom of Fu-nan, probably about 630 A.D. 

Having thus definitely laid down, in outline, the course of events 
by which the vassal state of Kambuja came to take the place of the 
old kingdom of Fu-nan, we may now proceed to discuss its history 
in some details. 

It is evident that there were two separate royal dynasties one 
founded by Srutavarman, and the other by Bhavavarman. The 
first dynasty at first acknowledged the suzerainty of Fu-nan, but 
seems later to have achieved the independence of the kingdom. 
But it was under the second dynasty that Kambuja became great 
and powerful and ultimately took the place of Fu-nan as suzerain 
of the whole of Cambodia. 

Of the two kings Srutavarman and Srcsthavarmart of the first 
dynasty we know but little. Nor do we know how they were con- 
nected with Bhavavarman who was the real founder of the second 
dynasty. That there was some connection may be presumed from 
the qualifying expression 'Srutavarma-mula' applied to the kings 
of Kambuja. This general reference to Srutavarman, as the com- 
mon ancestor of the Kambuja kings, is corroborated by the fact 
that even king Jayavarman VII, who ruled towards the close of 
the twelfth century A.D., claims to have been descended from Sres- 
thavarman, the supreme king of l&resthapura (No. 177, vv. 6-7) . 
King Udayadityavarman I is also said to have been descended 
through his mother from the family of Sresthapura (No. 117 v. 5) . 

The Ta Prohm Ins. (No. 177) seems to throw some light on the 
relation between the two dynasties. After referring to Sre^thavar- 


man in v. 7 it devotes the next verse to the eulogy of a lady, born 
in his maternal family (tadlye mdtr~kiil-amburas(iu) , She is called 
Kambu ja-raja-laksml. This has been taken as a proper name- 
but must be regarded as a very unusual one. It may be also taken 
as a descriptive epithet, though in that case we have to assume 
that the personal name of the lady is not mentioned. The next 
verse, i.e., v. 9, refers to Bhavavarman as 'bhartd bhuvo Bhavapure' 
and 'avamndra-kula-prasuteh kartta,'. These two expressions may 
mean no more than that Bhavavarman was the lord of Bhavapura 
and the founder of a royal family, but may also contain a veiled allu- 
sion to the fact that he was the husband of the lady. If some 
relationship did not sulr-i'rt between. Bhavavarman and this lady, it 
is difficult to account for the insertion of v. 8 between the eulogy 
of Sresthavarman in v. 7 and that of Bhavavarman in v. 9. If there 
was any relationship, the expressions referred to above would 
certainly favour the hypothesis that Bhavavarman was the husband 
of this lady. 

This hypothesis is supported by another consideration. The 
Ta Prohm Ins. (No. 177) compares 6resthavarman to the sun and 
Bhavavarman to the moon, while the Baksei Camkron Ins. (No. 89) 
claims that the lineage of Kambu Svayambhuva brought about the 
alliance of the solar and lunar races. A consideration of all the 
facts leads to a reasonable inference that the families of Srutavar- 
rnan and Bhavavarman were connected by marriage, even though 
we may not regard Bhavavarman as the husband of the lady called 

The original kingdom of Kambuja must, as noted above, have 
comprised the region of Bassac along the river Mekong where it 
emerges from Laos into Cambodia proper, the capital city, Srestha- 
pura, being situated near the modern sanctuary of Vat Phu. This 
was the early seat of the Kambuja rulers traditionally associated 
with Kambu Svayambhuva. With Bhavavarman, there is not only 
a change of capital city to Bhavapura, but also a break in that tradi- 
tion. Bhavavarman, Mahendravarman and their successors make 
no allusion either to Kambu or to Sresthapura, but describe them- 
selves, like kings of Fu-nan, as descendants of Kaundinya and 
Soma. It has been suggested 7 that this was a clever political move 
en their part in order to make themselves appear as the legitimate 
heirs of the kings of Fu-nan. Perhaps the same motive also im- 
pelled them to build numerous religious sanctuaries in the southern 

7. BEFEO. XVIII. No. 9. p. 3. 


regions. In any case it was not till the reign of Jayavarman II in 
the ninth century A.D. that the kings of Kambuja again refer to 
the old tradition of Kambu, gresthapura, Suryavamsa etc. As by 
that time the glory of Fu-nan was a thing of the remote past, and 
the political centre of Kambuja was shifted far to the north, it was 
perhaps no longer necessary to recall its traditions. 

This explanation is no doubt a plausible one, but does not 
carry immediate conviction. It is equally possible that Bhavavar- 
man was more intimately connected with the kings of Fu-nan than 
we yet know, and" that may be the reason why he and his successors 
carried on the old traditions of that kingdom. As noted above, he 
founded a new kingdom. Several records refer to his grandfather 
as 'sarvabhauma' Le., a suzerain or emperor, but there is no evi- 
dence that his father Viravarman was ever a reigning king. It 
has been suggested, with some degree of plausibility, that the suze- 
rain was no other than Rudravarman, king of Fu-nan, whose name 
heads the list of kings in the Ang Chumnik Ins. referred to above. 
According to this theory the death of Rudravarman was followed 
by a disputed succession between two princes, one living at Fu-nan, 
?nd the other, Bhavavarman, who got the kingdom of Kambuja aB 
his r.ppanage. Ultimately Bhavavarman got the better of his rival, 
and his successors finally extinguished the main line of kings at 
Fu-nan and ruled over the whole country. This theory 8 no doubt 
lacks definite proof, but certainly offers an explanation of the 
change of traditions which is hardly less plausible than the one 
mentioned above. It also offers a better explanation of v. 16 of 
Ins. No. 89 which refers to 'kings beginning with Rudravarman' as 
succeeding the royal line of Srutavarman. 

Without pursuing any further the speculative hypothesis re- 
garding the relation of Bhavavarman with the dynasty of 6rutavar- 
man of Kambuja on the one hand and the royal family of Fu-nan 
on the other, we may now proceed to reconstruct his history on 
the basis of epigraphic evidence. As regards his ancestry, it has 
been incidentally mentioned above that he was the son of Viravar- 
man and the grandson of Sarvabhauma, This we know from six 
records (No. 11) which are nearly identical, and commence with 
the following verse about Mahendravarman. 

Naptd sn-Sarvabhaumasya siiniis^sri-Viravarmanah 
$aktya~nilno kanistho<~pi bhrdtd srl-Bliavavarmandh \\ 

8. BEFEO. XXVIII, pp. 130-31. Melange Sylvatn Levi, pp. 210-11; 
BCAI, 1911, p. 36, 


It means that Mahendravarman was the grandson of Srl-Sarva- 
bhauma, the son of Viravarman, and the youngest brother of 
Bhavavarman, although not inferior to the last in prowess. That 
Bhavavarman was the son of Viravarman is also known from Ins. 
No. 9 which refers to the wife of the donor as a daughter of 
Viravarman and sister of Bhavavarman. It is worthy of remark 
that in both the cases Viravarman is mentioned without any royal 
title. This raises some doubts about the view, mentioned above, 
which seeks to interpret the word Sri Sarvabhauma as suzerain 
instead of taking it as a personal name of his grandfather. The 
prefix Sri seems to indicate it as a personal name, 9 and the context 
in which it occurs is undoubtedly in favour of this interpretation. 
On the other hand it must be admitted that this is rather an unusual 
name for a person (though names like Acarya Vidya-Vinaya 
(No. 22) are not less so), and the absence of a name-ending like 
Varman is also against this view. On the whole the question must 
be left open. 

The history of Bhavavarman's reign has been usually recon- 
structed on the basis of Inscriptions Nos, 5-9 all of which mention 
king Bhavavarman but do not contain any date. As no other king 
of this name was known until the discovery of Ins. No. 24, all the 
five inscriptions were referred to Bhavavarman, the founder of this 
dynasty, and very important conclusions were drawn from 
them about his life and reign. As usually happens, these have not 
been seriously questioned even after the discovery of the Ins. No. 24 
which definitely establishes the existence of a second king of that 
name who flourished more than a century after his namesake. Thus 
it is generally concluded from the findspots of these five inscriptions 
that Bhavavarman was a great conqueror and extended his king- 
dom up to the province of Battambang in the west (No. 7) . This 
would mean that he was master of nearly the whole of Cambodia 
with the possible exception of a small strip of territory to the south. 
This is hardly consistent with the Chinese -account which attributes 
to Mahendravarman and his son the conquest of the kingdom of 
Fu-nan which comprised at least a large part of these territories, 
if not the whole of them. It is no doubt possible to reconcile the 
Chinese account with the epigraphic data by arguing that Bhava- 
varman conquered all the outlying possessions of Fu-nan leaving 
to his successors the task of finally subjugating the small kingdom 

9. Coeds also holds this view (BEFEO, XXII, 58-59). 


of Fu-nan proper. But the fact remains that we have no right to 
assume that Bhavavarman I really exercised sovereignty over such 
a vast area. For some of the inscriptions at any rate might belong 
to the reign of Bhavavarman II As a matter of fact Ins. No. 9 
alone can be definitely assigned to the reign of Bhavavarman I, 
from the mention of his father's name, but the remaining four might 
belong to the reign of either of the two kings bearing the name 
Bhavavarman. Palaeography is not of much help in distinguishing 
the records of the two kings, for as Coedes has rightly pointed out, 
the essential features of the'Kambuja alphabet remained unaltered 
up to the time of Jayavarman I. Nevertheless he has suggested 
from certain characteristics in the style of writing that the Ins. 
Nos. 7 and 8 should be attributed to Bhavavarmau II, 10 and even 
this would take away the evidence for the southern limit of the 
kingdom of Bhavavarman I, referred to above. All that, therefore, 
we can definitely say is that Bhavavarman ruled in the north- 
eastern part of Cambodia where Ins. No. 9 was found. The inscrip- 
tions Nos, 5, 6, 8 describe the virtues and prowess of the king in 
the most extravagant manner and in a high-flown kavya, style in 
right Indian fashion. But beyond denoting the limits of the king- 
dom they yield very little historical infoiTnation. The Han Chey 
Ins. (No, 8) gives him the title Maharajadhiraja and says that he 
has violated the honour of the Aila race only inasmuch as he has, 
by his prowess, exceeded the limits of their territories (v. A. 17). 
It refers to his conquest of hill-forts (v. B. 5) and adds that 
'enemies, although not vanquished in battle, are attracted by his 
influence and bow down at his lotus feet with the offer of princely 
fortune' (v. A. 13). But as already noted above, it is doubtful 
whether the Han Chey Ins, is to be attributed to Bhavavarman I 
or Bhavavarman II. 

There is some uncertainty about the successor of Bhavavarman 
I. The Chinese evidence, and other epigraphic records, noted 
above, seem to suggest that Bhavavarman was succeeded by his 
brother Citrasena. On the other hand the Han Chey Ins. (No. 8) 
definitely asserts that the younger son of Bhavavarman peacefully 
ascended the throne of his father, and the donor mentioned in that 
record is said to have served under both these kings (v. 22) , This 
Bhavavarman has been taken to be Bhavavarman I, although, as no- 
ted above, the identity cannot be regarded as certain. But if we accept 

JO. BEFEO. IV, p, 


the identity we must hold that Bhavavarman's son had a short 
reign. It is impossible to say in that case whether he had a natural 
death at a young age or was killed by his uncle Citra&ena. That 
such a contingency was not very unlikely is proved by the follow- 
ing observation in a Chinese account which was recorded not very 
long after the accession of Citrasena: "The day a new king ascends 
the throne his brothers are mutilated by their nose or fingers being 
cut, and they are kept in confinement, each in a separate place". 

Citrasena Mahendravarman has left us two records. The first 
of these (No, 10) under the former name has three copies at Cruoy 
Anphil, Thma kre, and Tham Pet Thong. The second (No. 11) 
under the latter name is known from six copies (more or less exact) 
at Chan Nakhon (Phu Bahkon) , Khan Thcvada (2 copies) , Tham 
Prasat, Muang Surin, and near Keng Tana. It is expressly stated 
in the second that he was formerly known as Citrasena and assumed 
the name Mahendravarman at the time of coronation. It may 
therefore be argued that the first series of records were engraved 
before he ascended the throne, particularly as no royal epithet is 
given to him. But it must be pointed out that the Chinese call 
him Citrasena even after he became king. It may also be doubted 
if the findspots of these inscriptions were included within the king- 
dom. But it is certainly more reasonable to hold that they were, 
and on that hypothesis we may regard the kingdom of Kambuja 
as having been extended in the south along the Mekong valley up 
to Thma Kre beyond Sambor, and in the west along the Mun valley 
up to Tham Pet Thong near the sources of that river in Rajasima 
district in Siam beyond the Dangrek mountains. Whether these 
territories were conquered by Citrasena or his predecessors must 
be left an open question. The inscriptions of Mahendravarman 
show a further extension of his kingdom to the north along the 
valley of the Mekong up to the region of Chan Nakhon beyond 
Basac. But none of his inscriptions has been found to the west 
of M. Surin which lies considerably to the east of Tham Pet Thong. 
The Chinese accounts referred to above would indicate that he 
proceeded further south along the Mekong valley towards Fu-nan, 
and conquered the capital of that kingdom situated probably at 
Ba Phnom, but it is difficult to fix the limits of his conquest. 

The Chinese accounts also enable us to fix the date of Ma- 
hendravarman within narrow limits. His conquest of Fu-na<n, his 
death, and the accession of Isanavarman must all be placed during 
589 to 618 A.D., the period covered by the History of the Sui 
dynasty which refers to them. The same chronicle refers to an 


embassy from Kambuja in the year 616 or 617 A.D. 11 As the 
Chinese account was presumably based upon the report of this 
embassy, it must have been sent by Isanavarman. Mahendravar- 
man, therefore, must have died beiore 616 or 617 A.D. As Rudra- 
varman, the last known king of Fu-nan, certainly ruled from 517 
to 539 A.D. and Bhavavarman was removed by only one generation 
from him, the date of the two brothers Bhavavarman and 
Mahendravarman may be roughly placed between 550 and 600 A.D. 

The Ang Chumnik Ins. (No. 27) tells us that Mahendravarman 
sent an ambassador to the king of Champa for renewing the friend- 
ship between the two nations. It may be concluded, in the light 
of la,ter events, that Kambuja had deliberately begun to play a 
part in the politics of that neighbouring kingdom which bore such 
important results in the next reign. 

Isanavarman succeeded his father Mahendravarman about 600 
A.D. According to a Chinese text Isanavarman put his brothers 
in solitary confinement in order to put an end to the rivalry for 
the throne. 12 The main event of his reign was the protracted 
struggle with Fu-nan, for as noted above, Mahendravarman, in 
spite of his successes could i^ot finally subjugate the country, and 
its king, though forced to remove his capita/1 further south, still 
maintained a precarious existence in a corner of his kingdom. He, 
however, evidently offered stubborn resistance as it was not till 
627 A.D., and possibly a few years later, that Isanavarman finally 
extinguished the kingdom and took possession of its territory. 13 
The Chinese account is corroborated by the inscriptions of the 
king as they are found in the valley of the lower Mekong, both to 
the east and west of Chaudoc (Nos. 12 and 18) . There is no doubt 
that his kingdom extended along the valley of the Mekong from 
its junction with the Mun to its mouth. His kingdom must have 
comprised the whole of Cambodia and also the valley of the Mun 
to the north of the Dangrek mountains. The Vat Chakret Ins. 

11. BEFEO. Ill, p. 272. 

12. E. Aymonier Histoire de I'Ancien Cambodge, p, 32. I have not been 
able to verify this statement. It seems N to be based merely on the general 
observation, referred to above, in the History of the Sui Dynasty, which 
naturally may be taken as applicable to the case of Isanavarman whose em- 
bassy to China supplied the requisite information to the Chinese historian. 

13. According to the Ntw History of the T'ang Dynasty isanavarman con- 
quered Fu-nan at the beginning of the period 627-649. The same text refers 
to an embassy from Fu-nan during the same period (BEFEO. Ill, pp, 274-*75). 


(No, 21) dated in the year 549 Saka or 627 A.D. says that he, the 
lord of Tamrapura, possessed the three cities of Cakrankapura, 
Amoghapura, and Bhimapura. Of these the first has been identi- 
fied with Chikreng or Chakreng to the south-east of Angkor, the 
second with Battambang, and the third with Phimai on the Semun. 14 
An inscription found near Chantabun shows that the region on the 
border of Siam was included within the domains of Isanavarman. 15 
The epigraphic evidence supports the statement of Hiuen Tsang 
that Isanavarman's kingdom comprised the central part of Indo- 
China, with the kingdom of Dvaravati (Central Siam) on the west 
and Maha-Champa (Annam) on the east. 16 

Isanavarman transferred his capital to a city named after him 
Isanapura. This has been identified with Sambor Prei Kuk, 17 
where a large number of his inscriptions have been found. This 
was the beginning of that shifting of the seat of political authority 
towards the west which ultimately led to the establishment of the 
political centre of Kambuja in the Angkor region. 

As already noted above, Isanavarman sent an embassy to 
China in 616 or 617 A.D. Isanavarman's name is also intimately 
associated with the history of Champa. That kingdom was then 
passing through a series of palace revolutions and political intrigues 
of which the exact nature is difficult to determine. It is quite clear 
that the kings of Kambuja, specially Mahendravarman ajid his son, 
took an active part in its affairs. The daughter of Isanavarmam, 
princess Sri Sarvani, was married to Jagaddharma of Champa, and 
ultimately her son Prakasadharma secured the throne of Champa 
and restored order and tranquility. 18 

The list of kings preserved in the Ang Chumnik Ins. gives the 
name of Jayavarman after that of Isanavarman. The Ins. No. 24, 
however, reveals the existence of a king Bhavavarman reigning in 
561 Saka or 639 A.D. As Isanavarman was on the throne in 627 

14. Maspero l/?nptre Khmer, p. 27 fn. BEFEO. XXIV, pp. 359 ft. 

15. BEFEO. XXIV. pp. 352-358. It was held by Lajonquierrc that the 
Khmers did not extend their power to the Chantabun region till the 9th 
century A.D. But a fragmentary inscription at Chantabun containing the 
name of Isanavarman, and another at Khalung, in the same region, disprove 
this view. 

16. Beal Records, II. p. 200. 

17. BEFEO. XXVIH, p. 125. 

18. Champa pp. 39-45. 


AJD. (No. 20) and the earliest known date of Jayavarmaoi is 657 
A.D., the reign of Bhavavarman II may be placed between 635 and 
650 A.D, Unfortunately we know nothing of this king, not even 
his relationship with the two kings mentioned above. An inscrip- 
tion (No. 23) at Phnom Bayan mentions a donation by king Bhava- 
varman to Utpannesvara (probably a form of Siva) . This Bhava- 
varman has been identified with Bhavavarman II. The inscription 
contains a reference to Kaundinya and his queen at the beginning 
and mentions the 'descendants of the lunar race of . . . .Sri Kongar- 
varman'. This latter name is otherwise unknown in the history 
of Kambuja, but recalls the similar names of the Ganga kings of 
India. 19 

The next king known to us, Jayavarman I, is referred to in 
several inscriptions (Nos. 22, 25-27, 30, 30A, 32, 33, 34) which 
refer, in general terms, to his great prowess and conquests and 
manifold virtues. His relationship with the preceding kings is not 
known. The earliest known date of his reign is 657 A.D. (No. 25) 
and the latest 674 A.D, (Ins. No. 30A) . Ins. No. 26 refers to the 
transmission of certain properties by two Buddhist monks to the 
son of the daughter of their sister. For a long time it was believed 
to have been the earliest Buddhist inscription, but this has been 
proved to be wrong by the discovery of the Buddhist inscription 
of Rudravarman, the king of Fu-nan. The Ang Chumnik inscrip- 
tion dated 667 A.D. (No. 27) , which has been frequently referred 
to in the preceding pages and of which a summary has been given 
above, was a record set up by Simhadatta who was at first the royal 
physician, and then appointed the governor of Adhyapura by king 
Jayavarman L Jayavarman I is the last of the five or six kings 
who are known so far to have ruled over the kingdom founded by 
Bhavavarman I. The period covered by their reigns is about a, 
century. But it may be regarded as one of the most important for 
the history and culture of the Hindu colonial kingdom of Kambuja. 

In the case of Fu-nan we are chiefly dependent on the Chinese 
accounts, but as regards Kambuja, we have, in addition, the 
evidence of Inscriptions which throw a flood of light on the history 
and civilisation of the country. About one hundred inscriptions, 
belonging to this period are so far known to us, written both in 
Sanskrit and the native language Khmer. We shall now try with 
the help of these two-fold sources of information to reconstruct a 

19. Cf. JGIS. V. 156. 


picture of the Hindu colony of Kambuja in the seventh century 

We have seen how under a line of able rulers the small vassal 
state of Kambuja rapidly grew to be a powerful kingdom. It not 
only established its authority over the whole of Cambodia and 
Cochin-China, but a considerable portion of modern Siam was 
comprised in it. The whole of the valley of the Mun river to the 
north of the Dangrek mountains was ruled by the Kambuja kings, 
and an imaginary line drawn due north from Chantabun to the 
source of that river represents the western limit of their kingdom 
as testified to by the find-spots of inscriptions and testimony of 
Hiuen Tsang. Part of these territories was, as it still is, inhabited 
by primitive savage tribes, and rendered difficult of access 
on account of hills and dense forest. We get a reference to it 
in Ins. No. 32 which records the career of an officer under king 
Jayuvarman. This officer was appointed first as Mahasvapati, 
Master of the Horse, and then as governor of Sre^lhapura (near 
Bassac in Laos) and Dhruv^pura. We are told that he kept 
Dhruvapura free from troubles, although it was full of dense terri- 
ble forest (bhisaTWiranyorsankatam) and peopled by ferocious 
tribes (uddrpta-pimis-^dvdsa) . Reference is also made to the 
defeat inflicted upon mountain chiefs (parvata^-bhiipala) in Ins. 
No. 8. 

These instances 20 serve to show how, starting from the Mekong 
valley as the centre, the Hindu rulers of Kambuja gradually 
extended their power and authority towards the inaccessible regions 
in the north and west and gradually brought them within the 
sphere of Hindu culture and civilisation 

The Hindu colonists in Kambuja set up an administrative 
system on Indian model. The study of Arthasastra is specially 
mentioned in an inscription (No. 27, v. 6) and most probably 
Kautilya's Arthasastra or some text of this type was actually 
followed in practice. An interesting evidence of this is supplied 
by a passing reference in Ins. No. 8. The last verse of this inscrip- 
tion says about the donor, a royal official, that he was a favourite 

20. V. 7 of the Ins. No. 7 also refers to dense forest frequented by tigers, 
evidently in reference to a territory conquered by Bhavavarman, but as the 
second line of the verse is missing, no complete sense can be made of the 


of the king (antarangatvam-dsthitah) as he was sarvopadha- 
suddha. Earth, who edited this inscription, failed to realise its 
meaning and recognise its importance. The term is, however, used 
in Kautilya's Arthasastra (I. 10) in connection with the testing of 
the conduct of royal officials by four kinds of allurements, which 
are technically known as upadlws. One who does not fall a prey 
to any of these allurements, to which he is tempted by royal spies, 
and thereby successfully passes all the tests, is known cs mrvo- 
padhdsuddha, and Kautliya recommends that he should be 
appointed a minister. The occurrence of this word in connection 
with a royal servant therefore indicates that probably the method 
recommended by Kautilya was followed by Kambuja kings in the 
appointment of ministers. 

Unfortunately the inscriptions do not enable us to form a clear 
idea of the system of administration as a whole, and only give us 
a few glimpses of it. The king's authority was supreme and a 
divine origin was claimed for him (No. 32, v. 3), The mention 
of an officer called Rajasabhadhipati (No. 30, v. 6) indicates the 
existence of a royal assembly, but we have no knowledge of its 
nature and constitution. The 'minister' is also frequently referred 
to and held an important position in state. Several high officials 
are mentioned in the inscriptions, but it is not always easy to 
distinguish their exact status and nature of duties. The duta or 
ambassador is referred to in one record. Another high officer 
seems to have been something like the Chief of the royal household 
who was in charge of the royal insignia, particularly the valuable 
ornaments of the king (No. 32 v. 18). The royal physician was 
r,n important personage as Ins. 27 shows. 

Several military offices are referred to. The Mahasvapati, as 
the name indicates, was probably the commander of cavalry. 
Another highly honourable post was that of the leader of the royal 
body-guard (nrp^&ntaranga-yaudha) composed of soldiers with 
arms and helmets. It is perhaps this body which is referred to in 
the following passage in a Chinese Chronicle: "In front of the 
chamber containing the royal throne there are thousand guards 
armed with cuirass and lances' 5 . Two naval officers are men- 
tioned, maha-nauvahdka (No. 45) and Samanta-nauvdha (No. 32, 
v. 18) . The latter is said to be the head of the boatmen (Taritra- 
bhri} vh^ knows their classification. 

The inscriptions frequently refer to war elephants. Accord- 
ing to a statement in the History of the Tang dynasty which appears 


from the context to refer to the 7th century A.D., there were five 
thousand war-elephants in Kambuja. 

There seem to be different gradations of military rank. The 
commander of a body of troops living in a single city is called 
Sahasra-Vargg-adhipatih (No. 32, v. 19);. The word 'vargga' 
evidently stands for a unit, and the title probably means the com- 
mander of a unit of thousand soldiers, leather than one of thousand 

The kingdom was divided into a large number of districts each 
with a governor living in a city as his headquarter. The same 
Chinese chronicle tells us that there were thirty towns with several 
thousands of houses in each, and every town had a governor. 
Evidently the kingdom was divided into thirty administrative units, 
each with a headquarter at a city. The inscriptions not only con- 
firm this but supply us with names of a number of such towns 
viz. Pasenga (?), Tatandarapura, Tamrapura, Adhyapura, 
Sresthapura, Bhavapura, Dhruvapura, Dhanvipura, Jyesthapura, 
Vikramapura, Ugrapura, and Isanapura. 

The towns were surrounded by walls and ditches. Among the 
amenities of life provided therein are mentioned public institu- 
tions like Viprasdld, Sarasvati (Public school or library?), satra 

(guest house or hospital) , and tanks, both big and small (No. 32, 
v. 8), bhaktasdld (alms-house) and a silayandhwna (bridge?), 

(No. 46). 

A pointed reference to Isanavarman as suzerain of three kings 
and lord of three cities perhaps indicates that there were three 
vassal states (No. 19) not directly administered by royal officials 
but enjoying some sort of internal autonomy. These three cities 
were probably Cakrankapura, Amoghapura and Bhimapura 
(No. 21) mentioned above. 

As in ancient India the posts of minister and other high officials 
were often hereditary, the most notable being the case recorded in 
Ins. No. 27 to which reference has been made above. The gover- 
nors, ministers and other high officials received marks of distinction 
from the sovereign by the gift of a golden car and umbrella with 
golden embroidery, golden vessels (karanka-kalasa) , horses, ele- 
phants, retinue, etc. 

The court of Kaimbuja, which may be said to have developed 
into an empire, introduced pomp and grandeur befitting its power 


and glory. The contemporary Chinese chronicles 21 give a very 
interesting account of the court and capital of Isanavarman which 
is quoted below. 

"After Citrasena's death his son Yi-che-na-sian-tai (Isana- 
sena or Isanavarman) succeeded him. He lived in a town named 
Yi-chen-na (Isanapura). This town contained 20,000 houses. At 
the centre was a grand palace where the king held his court. 
Three days a week, the king sits in the court, on a seat decorated 
with five kinds of aromatics and seven kinds of precious stones. A 
costly canopy like a pavilion is placed over his head. Its columns are 
of painted wood. The walls are decorated with ivory and flowers of 
gold. The pavilion looks like a small hanging palace, all shining 
with gold. Two bowls of gold with the aromatics are carried by two 
men on two sides of the king. His crown is decorated with pearls 
and precious stones. His shoes are made of skin of different 
colours and decorated with ivory. He wears golden ear-rings and 
is always dressed in white. There are five kinds of high officers 
and they are dressed like the king. The officers touch the ground 
three times in front of the throne of the king. The king then 
commands them to ascend the stairs and having done so they kneel 
down before the king with folded hands. They are then seated in a 
circle round the king for discussing state affairs. When the 
deliberation is over they kneel down, again prostrate themselves 
and go out. 

In contrast to the meagre information about administrative 
system, we possess a fair knowledge about the religion of Kambuja. 
Most of the inscriptions begin with an invocation to one or more 
gods and record donations to religious establishments. Taken as 
a whole they clearly indicate the strong hold of Indian religion on 
the population. Although Vedic sacrifices are referred to (Nos. 19, 
20), the Puranic form of religion, specially the worship of Si' r a, 
Visnu and the deities associated with them, was undoubtedly more 
predominant. The most popular god seems to be Siva though the 
composite deity Siva-Visnu, designated as Sankara-Narayana, 
Sambhu-Visnu (linga) , Har-achyuta, Hari-Sankara etc,, was also in 
great favour. Siva is occasionally referred to as the greatest 
god, whose feet are worshipped by Brhama and Visnu. 
He is described as a great ascetic and is known by various names 

21. History of the Sid Dynasty see f.n. 4 above. 


such as Amratakesvara, Rudra, Vyomesvara, Gambhlre6vara, 
Nikamesvara, Pihgalesvara, Naimisesvara, Isana, Srivijayesvara, 
Kedaresvara, Girlsa, gambhu, Tryambaka, Siddhesa, Trisuli, 
Sankara, Tribhuvanesvara, Nrttesvaxa, Acalesvara, Kadambesvar*, 
Mahesvara, Utpannesvara or Uppannakesvara, and was mostly 
worshipped in his Zmga-form. He bore Ganges on the forehead 
and had moon as his crest jewel His Bull (Nandin) was also 
regarded as sacred. The goddesses of the Hindu pantheon are not 
unknown and we have references to Uma, Durga Devlcaturbhuja, 
Bhagavatl, Laksmi and Sarasvati. Of the various names and forms 
of Visnu we come across Trailokyasara, Hari, Acyuta and Nara- 
yana. Reference is also made to fealagrama in a record by engrav- 
ing a figure of it after the name of the donor Salagramasvarm, 

Among other deities may be mentioned Yama, Prahantesvara, 
Candrayananatha, Tilakesvara, Mulasthana, Yajnapatisvara, Gana- 
pati and Svayambhu (Brahma) . Some of these may be names of 
Siva or Visnu. Buddhism was also prevalent, though judging by the 
number of records its influence does not seem to have been very 
great. According to the Chinese chronicle, however, there were 
many followers of Buddhism. One inscription refers to three 
Bodhisatvas, Sasta, Maitreya and Avalokitesvara. The epigraphic 
evidence is corroborated by the cult images actually discovered in 
Cambodia, among which mention may be made of Hari-Hara, Siva, 
Parvati, Nandin (Bull of Siva) , Brahma and Buddha. The images 
of the deities were placed in temples ruins of which lie scattered 
all over the country. The temple, near the capital, on the top of 
a hill, enjoyed special sanctity, and according to Chinese chronicle 
was guarded by five thousand soldiers. As in India, kings, high 
officials and even private persons vied with one another in setting 
up divine images, building temples for them and making endow- 
ments for the regular performance of their worship. The endow- 
ments generally consisted of gold, village, land (paddy fields), 
orchard, servants (both male and female, generally slaves) , beauti- 
ful women, probably devaddsis (No. 32, v. 22), cattle (cow, 
buffalo), arecanut and cocoanut trees. There were festivals of 
citizens in honour of 6iva (No. 22). Frequent reference is made 
to dsramas and we find already in the seventh century A.D. the 
beginning of those regulations which, multiplied and codified at a 
later age, form such a distinctive characteristic of Kambuja 
sanctuaries. It was, for example, laid down in an edict of Jaya- 
varman (No. 33) regarding the temple on Lingaparvata (Vat Phu) 
that no one living there, even if guilty, should be killed, that no 


one should go wherever he likes in this dsrama of god, use a car, 
an umbrella or a fly whisk, or bring dogs and fowls, 

The study of sacred literature is referred to in many inscrip- 
tions and we hear of Brahmanas proficient in Veda, Vedanga, 
Samaveda, and Buddhist scriptures, and even ministers with a 
profound knowledge of Dharmasastra. 

Reference is made to the daily recitation of the Ramayana, 
the Mahabharata and the Puraiias (No. 9). The inscriptions 
reveal a thorough acquaintance of their authors with these works. 
Ins. No. 36 refers to the gift of a manuscript of Sambhava, a work 
of Vyasa. There is no doubt that it refers to the section of Adi- 
parva of Mahabharata, called Sambhavaparva. Imprecations are 
invoked against anybody who destroys the manuscript deposited 
in the temple. 

Some of the invocations to God indicate that along with 
popular forms of worship religious philosophy of India was culti- 
vated to a high degree. Reference is made to the attainment of 
'Brahmapada,' or Niruttam Brahma as spiritual goal of life. Promi- 
nence is given to the all-pervading character of Siva whose eight 
bodies (asta tanu) consist of the moon, sun, sky, air, the atman, 
earth, water and fire. The invocation to Siva in Ins, No. 4 illus- 
trates Vedantic Saivism in which Siva is identified with Paramdt- 
man, or the Absolute of the Upanishads. The spiritual doctrine of 
the cessation of desire for the fruit of karma (action) as laid down 
in the Bhagavadgita is also echoed in it. 

A short record Om Jaiminaye svahd, 22 shows the great rever- 
ence for Purva-Mlmamsa, and probably indicates that its author 
was regarded as a deity and regularly worshipped. 

Some peculiar religious ideas of India are also met with. It 
is said of a royal official that he gratified the gods by Sivayajna, 
the ascetics by study, and the manes by the water offered by good 
sons (No. 27, v. 23). Reference is also made to ftrthas (holy 
places) and merits of pilgrimage. The Saiva Pasupata sect evi- 
dently wielded great influence, and an inscription refers to 
its dcdrya Vidyapuspa as a poet and versed in various branches 
of philosophy (No. 6). 

22. BEFEO. XXVIII, p. 43, 


Some debased forms of Saiva religion were also prevalent. 
According to the Chinese chronicle there was a temple to the east 
of the capital city, guarded by thousand soldiers, where the king 
went every year to sacrifice a human being. The Puraiiic myths 
and legends were also very familiar and the inscriptions abound in 
allusions to them. There are references to the churning of the 
ocean; the Kailasa; the Udayagiri; Kartika as general of the gods; 
dharma. crippled in Kaliyuga; the burning of the cupid; sacred 
character of tlrthas (including hills) ; lunar and solar races; Arun- 
dhatl, the ideal wife; Sesanaga, supporting the earth; Indra, the 
wielder of thunderbolt, and with thousand eyes, clipping the wings 
of mountains and performing hundred sacrifices (No. 20) ; Manu 
the first king; Dillpa as an ideal king; Asvini as the divine physician 
etc., etc. 

The secular literature was also regularly studied, and the ins- 
criptions refer to many of its branches such as Sabda, Vaisesika, 
Nyaya, Samiksa (Sankhya?) and Arihasastra. The Kavya formed 
a favourite subject of study, and even a minister is described 
(No. 27) as having drunk the nectar of poetry (a pita- 
kavita-rasah) , 

The Saka era was exclusively used and the expressions denot- 
ing date show a thorough acquaintance with the astronomical 
system of India. Reference is made to a governor, proficient m 
astrology (bhojakapravara No. 20). 

The inscriptions themselves furnish the best testimony to the 
assiduous cultivation of Sanskrit language and literature in Kam~ 
buja. Many of them are fairly long poems written in rich kdvya 
style and show high proficiency in the knowledge of Sanskrit 
grammar, vocabulary, idiom, rhetoric, prosody, metre and poetic 
conventions and styles. That the classical works in Sanskrit were 
regularly studied is evident from the use of familiar similes, com- 
parisons and allegories in the inscriptions, and it would be difficult 
to distinguish them from the prasastis composed in India. Judged 
by any standard, the writings of the Kambuja scholars must be 
regarded as no mean contribution to Sanskrit literature, 

As regards society, Indian institutions exercised great influence 
but were partially modified by indigenous ideas and customs. Of 
the caste divisions we find mention only of BrahmarjLas and 
K$atriyas. Reference is made to Brahmanas who were proficient 
in Veda and Vedangas and whose family for generations served as 
hotars (sacrificial priests). But even members of this family, as 


also of others, served in various offices of state, both civil and mili- 
tary. There thus seems to have been no rigidity of caste-rule as 
regards occupation. Inter-marriage between Brahmana and 
Ksatriya seems to have been a normal custom. The sister of king 
Bhavavarman, for example, was married to a Brahmana named 
Somasarman, proficient in Samaveda, and the issue of this marriage 
was Hiranya-varman, indicating the use of the epithet Varman in 
a Brahmana family. The Brahma-ksatra vaihsa is referred to in 
an inscription of the eighth century A,D. (No. 49), 

The reference to Bhavavarman's sister as 'pativrata* and 
dharmaratd (devoted to husband and religion) like a second Arun- 
dhati indicates that the high Indian ideal of womanhood was carried 
to these colonies. The great prominence given to sister's sons in 
several inscriptions (Nos. 27, 30) seems to indicate the prevalence 
of matriarchy, though nothing can be definitely asserted on this 
point. 23 

Unfortunately the inscriptions do not throw much light on the 
life of the common people. The Chinese chronicle, however, gives 
us a lot of interesting information about it, The following extracts 
from the History of the Sui dynasty describe the manners and 
customs of the people of Kambuja at the beginning of the seventh 

[Marriage]: 'They present the bride a robe. The families of 
bridegroom and bride stay for 8 days at home and keep lamps 
burning day and night. After marriage the husband takes a portion 
of his ancestral property and lives in a separate house with his wife'. 

[Funeral ceremony]: 'The children of the deceased do not 
eat or shave for seven days and utter loud cries. The relations, 
with the priests and priestesses, carry the dead with prayers and 
music, burn the body with all kinds of aromatic woods, put the 
ashes in an urn of gold and silver and throw it in a big river. The 
poor use earthen jar painted in different colours. Sometimes they 
do not burn the body but leave it in the hills to be devoured by 

[Epidemics]: 'To prevent epidemics, sacrifices are offered 
beyond the western gates of the town by killing pigs, bulls and lambs 

23. Later inscriptions of Kambuja refer more directly to the succession 
through females, a custom still prevalent in Laos (Corpus, pp. 124-126, 179-80). 


of white colour. They believe that otherwise grains will not ripen, 
domestic animals will die and large number of people would fall 
victim to epidemic'. 

[General nature and habits]: 'The men are short and have a 
dark complexion. But there are white women. The people dress 
their hair and wear ear-rings, They consider the right hand as 
pure and the left impure. Every morning they make ablutions, 
cleanse teeth by small pieces of branches of trees, read books, say 
prayers, again make ablution, take food, cleanse teeth after meal, 
and again say prayers. Their food includes a large quantity of 
butter, cream, sugar and millet (in the form of cake or bread) . 
Before meal they take some morsels of roasted meat with bread, 
which they eat with a little salt'. 

The following statement about Kambuja, in the History of the 
T'ang dynasty, probably refers to the seventh century: 

'The houses are all turned towards the east. They welcome 
the guests with areca, camphor and perfumes, for they do not drink 
publicly but only with their own wives at home avoiding the 
presence of parents'. 

In spite of the fragmentary nature of the information we have 
culled above from the Chinese chronicles and epi graphic evidence, 
the dominance of Indian influence in the development of culture 
and civilisation in Kambuja is clearly manifest. This is only what 
could be expected in a country colonised by the Indians. It may 
be safely presumed that they maintained regular contact with their 
motherland. The king Isanavarman, for example, is specifically 
referred to in an Inscription (No, 16) to have relations with India. 
But the most striking evidence of a continuous contact with India 
and of her serving as the perennial spring which fed the fountain 
of Indian culture in Kambuja is furnished by the development of 
Indian art in that far-off colony, to which a brief reference has 
already been made in the last lecture, 




The seventh century A.D. witnessed the rise of Kambuja as a 
great political power and a flourishing centre of Hindu culture and 
civilisation. But its history during the eighth century A.D. is 
shrouded in darkness and obscurity. A few isolated epigraphic 
data and a brief account preserved in the Chinese chronicles enable 
us to form a very vague idea of its general condition, but it is 
impossible to give any connected outline of its political history 
during the century following the reign of Jayavarman I, the last 
known king who ruled over the kingdom founded by Bhavavarman 
i. The only certain information that we may derive is that the 
mighty and extensive kingdom over which the dynasty of Bhavavar- 
man ruled had been divided into a number of states, and we possess 
the names of a few of them and some of their rulers. But the location 
of these states and the names and order of succession of their rulers, 
far less their activities, cannot be definitely ascertained. The eighth 
century A.D. may thus be justly described as the dark period of 
the history of Kambuja, fortunately the only dark period in its 
almost unchequered history of thousand years. All that we can 
do, so far as this period is concerned, is to bring together such 
evidences as we possess and try to correlate them as far as available 
materials permit. 

According to the Chinese annals of the Tang Dynasty, shortly 
after 705-706 A.D., i.e. at the beginning of the eighth century A.D., 
Chen-la or Kambuja was divided into two states, viz., Kambuja 
of the land and Kambuja of the water. The former, called also 
Wen-tan or Po-leu, comprised the northern part of Cambodia, full 
of hills and valleys, and the latter covered the southern part which 
bordered on the sea and abounded in lakes and streams. 1 Ma- 
Tuan-Lin confirms this account and adds that Chen-la or Kambuja 
of the water had an extent of 800 li and that its king 
inhabited the town of Po-lo-ti-pa. 2 Until recently the 

1. BEFEO. XXXVI, p. l. 

2. Ibid, p. 5. 


Chinese account was taken to mean that the Kambuja pro- 
per was divided into two kingdoms, and attempts were made 
by various scholars to define the boundaries of, or at least to 
locate, these two kingdoms which the Chinese so characteristi- 
cally referred to as Kambuja of land and Kambuja of water. The 
fact, however, appears to be that by Chen-la or Kambuja of the 
land the Chinese referred to a kingdom to the north of Kambifj!? 
proper, including a great part of Laos and touching the Chinese 
province of Tonkin and the Thai kingdom of Yunnan. This 
kingdom maintained diplomatic relations with China and sent an 
embassy to the Imperial court in 717 A.D. But five years later 
we find this kingdom sending an army to help Mei Hiuan-Cheng, 
the frontier chief of Nghe-an in Annam who had revolted against 
the Chinese emperor and was joined by several other chiefs of 
hilly regions. The Kambuja army joined the rebel chief of Annam 
and defeated the Chinese forces. 2a This incident shows that the 
northern Kambuja kingdom was a fairly powerful one. 

The friendly relations of the kingdom with China were, how- 
ever, soon restored, and an embassy was sent in 750 A.D. In 753 
A.D. the son of the king visited the Chinese Court with a retinue 
of 26 persons and accompanied the Chinese military expedition 
against Nan-chao. In 771 the king Po-mi paid a visit to the 
Chinese Emperor. The last embassy to China was sent in 799 
A.D. In spite of these frequent references, the Chinese accounts 
do not enable us to precisely determine the location of the king- 
dom. It appears, however, from the itinerary of Kia Tan, that in 
the eighth century the nominal suzerainty of China was extended 
as far as Labs, and Wen-tan, or Kambuja of the land, touched 
the Chinese province of Tonkin. This indicates that Wen-tan 
extended along the middle course of the Mekong. 3 

As regards Chen-la of water or the Kambuja proper we know 
the names of a few kingdoms and their rulers from the inscriptions 
of king Yasovarman who flourished towards the close of the ninth 
rentury A.D. These records begin with a genealogical account of 
the king the first part of which may be summed up as follows: 4 

"There was a descendant of the lord of Aninditapura, Sri 
Puskaraksa by name, who obtained the kingdom of Sambhupura. 

2a. BEFEO. XVIII, No. 3, pp. 29-30. 

3. BEFEO, IV. pp. 211-12, 

4. Corpus, p. 364, 


Rajenrlravarman, who was born in the family of this king and 
whose mother was descended from the suzerains (adhirdja) of 
Vyadhapura, also ruled in Sambhupura". 

The first part of this genealogy is further elucidated by that 
of Rajendravarman given in Pre-Rup Ins. (No. 93) and Mebon 
Ins. (No. 89A). According to these, Puskaraksa was the son of 
king Nrpatlndravarman who was descended from Sarasvati, the 
sister's daughter of Baladitya, king of Aninditapura and a descen- 
dant of Kaundinya and Soma. These relationships may be shown 
by the following genealogical table: 5 

Kaundinya= Soma 

Baladitya D 




Of the three kingdoms mentioned in this extract the location of 
Sambhupura admits of no doubt. It is represented by modern 
Sambor on the Mekong, 6 

As regards Vyadhapura, Aymonier's identification of it with 
Ankor Borei in the province of Prei Krebas held the ground till 
Coedes demonstrated that it was more probably situated at the 
foot of the hill called Ba Phnom and possibly this kingdom repre- 
sented that of ancient Fu-nan. 7 

The site of Aninditapura is not definitely known, but according 
to Coedes it must be looked for in the region east of Angkor on 
the north side of the Great Lake (Tonle Sap). 8 

The somewhat curious and pompous genealogy of Yasovarman, 
recorded at length in identical words in a large number of his 
inscriptions (Nos. 60-63), is an elaborate attempt to connect that 
ruler, however remotely, with the three ruling families of 
Kambuja. Whatever historical truth there may be in this genea- 

5. Inscriptions, pp. 74-75. 

6. Aymonier, I, p. 309. 

7. BEFEO. XXVin. pp. 127-131, 

8. Ibid, p, 133. 


logical pedigree, these official documents leave no doubt that to- 
wards the close of the ninth century A.D., when they were drawn 
up, the three royal families of old times occupied a position of 
eminence in the memory of the people. 

The first question, therefore, that arises in this connection is 
whether there is any independent evidence that Sambhupura, 
Aninditapura and Vyadhapura flourished as independent kingdoms 
in the eighth century A. D. 

Following the interpretation of Yasovarman's records by Ber- 
gaigne it is generally held thai Rajendravarman, the great-grand- 
father of Yaovarman, married the heiress of the kingdom of 
Vyadhapura and thus added that kingdom to his own. But this 
interpretation is not justified by the language of the records which 
merely says that Raj endravar man's wife was descended from the 
suzerains of Vyadhapura. 9 If we uphold the identification of 
Vyadhapura with the old kingdom of Fu-nan, the reference to it in 
Yasovarman's genealogy might mean no more than a claim to have 
been descended from the old royal family of Fu-nan which ruled 
over Cambodia till the middle of the sixth century A.D. Further, 
we should remember that the term udhiraja (suzerain), applied 
to the rulers of Vyadhapura, would bo more appropriate to the 
king of old Fu-nan, than to the local ruler of a petty state. It is, 
therefore, difficult to admit the existence of Vyadhapura as a 
kingdom in the eighth century A.D. on the basis of evidence so far 
available to us. 10 

We are perhaps more fortunate in the cases of the other two 
kingdoms. An inscription (No. 48) discovered at Prah That 
Kvan Pir, and dated in 638 Saka (=716 A.D,), states that one 
Puskara had a god Puskaresa consecrated by the ascetics and 
Brahmans. This Puskara has been identified with Puskaraksa, 11 
mentioned in the genealogical accounts of Yasovarman and 
Ilajendravarman, who was a descendant of Baladitya, king of 
Aninditapura, and obtained (probably by marriage) the kingdom 
of Sambhupura. The existence of these two kingdoms in the 
eighth century A.D., and even earlier, may thus be provisionally 
admitted. If the genealogical account of Yasovarman is to be 
fully believed, we must hold that these kingdoms were united, 

9. Ibid, p. 126 .n. (2). 

10. Ibid, p. 130. 

11. BEFEO. IV, p. 214. 


temporarily or permanently we cannot say, in the first half of the 
eighth century A.D. 

Three inscriptions 12 found in Cochin-China throw further light 
on this question. One of them, found at Thap Musi, refers to the 
installation of (an image or temple of) god Srl-Puskarak^a by 
king Sambhuvarman. The temple of Puskaraksa is also referred 
to in the second inscription found in the same place which records 
the installation of an image of Puspavatasvami in the sanctuary of 
Mulasthana. The third inscription found at the foot of the hill 
Nui Ba-the, in the district of Lon-Xuyen, refers to the construction 
of a brick temple on the top of the hill for the Vardhamana-Linga 
(of 6iva) for increasing the religious merit of king Sri Nrpaditya. 

Now, there can be no question that both Sambhuvarman and 
Nrpaditya, whose records have been found in Cochin-China, ruled 
in what the Chinese call Kambuja of the water. The name-ending 
dditya of the second king, in contrast to the varman, almost 
universally found in Kambuja, may indicate his association with 
Baladitya king of Aninditapura, referred to above. If we presume 
^uch a connection, the kingdom of Aninditapura may be regarded 
as corresponding to the Kambuja of the water. Po-lo-ti-po, the 
capital of the latter, according to the Chinese Annalb, may then be 
restored as Baladityapura after the name of Baladitya. But what- 
ever one might think of this, it is less difficult to connect gambhu- 
varman with Sambhupura, on the analogy of other towns named 
after the king, such as Bhavapura, gresthapura, Isanapura, etc. 

The fact that inscriptions of both Sambhuvarman and Nrpa- 
ditya are found in the same region naturally connect these two 
kings, and if we accept the hypothesis that the latter was connected 
with Baladitya, we may find here a confirmation of the statement 
in the records of Yaso varman, referred to above, according to 
which Puskaraksa united the two kingdoms of Aninditapura and 
Sambhupura. Further, it is interesting to note that two of the 
three inscriptions in Cochin-China referred to above mention the 
God Puskaraksa, who may after all owe this designation to the 
king of that name, who would thus be closely related to the other 
two kings, as suggested in the records of Yasovarman. 

Thus although no definite conclusion is possible, we may 
accept, as a provisional hypothesis, that shortly after the death of 

12. For the text of these inscriptions and the inferences drawn from them 
cf. BEFEO. XXXVT nn * ff 


Jayavarman I, Kambuja was split up into two kingdoms with 
Sambhupura and Aninditapura respectively as their capitals. The 
fact that the rulers of Aninditapura regarded themselves as des- 
cended from Kaunclmya and Soma, shows that the old traditions 
of Fu-nan, carried over by Bhavavarman and his successors, were 
still continued, and for all we know there might have been some sort 
of relationship between the royal family of Aninditapura and that 
of Bhavavarman, 

Unfortunately we possess no detailed account of any of the two 
kingdoms of Sambhupura and Aninditapura. In addition to the 
names of kings, mentioned above, who ruled over these kingdoms, 
in the eighth century A.D., the records of Yasovarman and his 
father Indravarman furnish names of other kings as will be noted 
in connection with their history, 13 But when and where they ruled, 
it is difficult to say. The names of a few other kings are supplied 
by a Khmer inscription dated 725 S. (803 A.D.) (No. 50). It re- 
fers to a religious endowment by the queen Jyestharya, and names, 
as her ancestors, the king Jayendra, the queen Nrpendradevi and 
the king Sri Indraloka. As the inscription is engraved on a temple 
at Sambor, these tailers may be regarded as kings of Sambhupura. 

Attempts have been made to identify the two kingdoms of 
Kambuja, referred to in the Chinese annals, with those mentioned 
in epigraphic records. Until recently it was generally held that 
the kingdom of Sambhupura corresponded to the Kambuja of 
land, and that of Vyadhapura, to the Kambuja of water of the 
Chinese chronicles. Coedes, however, dissented from this view 
and at first identified the last with Aninditapura. He now holds 
that the Kambuja of water more probably corresponded to the 
kingdom of Aninditapura, united with that of Sambhupura, while 
Kambuja of land denoted the territory north of Dangrek moun- 
tains. 14 

Whatever we might think of these theories, there is no doubt 
about the fact that during the eighth century A.D. Kambuja had 
lost the unity and solidarity which were imparted to it by the con- 
quests of Bhavavarman and his successors, and was split up into 
two or possibly more states, none of which was evidently of any 
considerable power or importance. Such a state of things could, 
of course, have been brought about by natural causes. But 
indications are not wanting that the fate of Kambuja was at least 
partly determined by external events. 

13. See pp. 91-92. 

14. BEFEO. XXXVI, 1 ft. where the older views, even other than those 
mentioned in the text, are given with references. 


The most outstanding fact in the political history of Indo-China 
and Indonesia in the eighth century A.D. is the rise of the Sailen- 
dras as a great power. 15 Their empire included Sumatra, Java, 
Malay Peninsula and a large number of islands in the Indian archi- 
pelago. There is no doubt that the northern part of Malay Penin- 
sula constituted a stronghold of their power and thus they were too 
dangerously near the western frontier of the Kambuja kingdom. 
Although positive evidence is lacking, there are reasonable grounds 
to suppose that the Sailendras extended their supremacy over 
Kambuja. In any case there is no doubt that Kambuja was a 
vassal state of Java towards the close of the eighth century A.D. 
when that island itself was at least partially conquered by the 

As a matter of fact we can trace Javanese influence ovei 
Kambuja from the beginning of the eighth century A.D. King 
Safijaya of Java is mentioned in his inscription, dated 732 A.D., as 
a conqueror of the countries of neighbouring kings. This some- 
what vague statement is corroborated by a detailed list of con- 
quests of the king in a literary work. After mentioning conquests 
in Java and Bali islands, this text refers to his over-seas expedition 
in course of which he proceeded to the Malaya country and fought 
with Kemir and other powers. 16 There is no doubt that Kemir 
stands here for the Khmers or the people of Kambuja. Ordinarily 
the value of such references may be discounted to a certain extent 
in view of the well-known tendency of the court-poets to exagge- 
rate the achievements of their royal patrons. But in this parti- 
cular instance the record of Kambuja itself supports the theory of 
a Javanese conquest of Kambuja. For, an inscription in Kambuja 
(No. 151) which we shall have occasion to discuss in details later, 
refers to a Kambuja king, who ruled towards the beginning of tho 
ninth century A.D., as having come from Java and performed a 
religious ceremony in order that Kambuja might not again be 
dependent on Java. 

The dependence of Kambuja on Java during the latter half of 
1he eighth century A.D. is also indirectly supported by the refer- 
ence in the inscriptions of Champa to Javanese naval raids on the 
coast of Annam as far as Tonkin in the north. An inscription 
dated 784 A.D. says that in 774 A.D. ferocious people of other 

15. Suvaruadvipa, I, Bk. II. Ch. I, 

16. Ibid, p. 230. 


cities came in ships and burnt a temple of Siva at Kauthara 
(S. Annam) and carried the Mukhalinga of the god. Another 
inscription dated 799 A.D. states that a temple was burnt by the 
army of Java coming by means of ships and became empty in the 
aka year 709 (787 A.D.). The Chinese annals also refer to an in- 
vasion of the northern part of Annam by the people of Daba, 
which Maspero identifies with Java, in 767 A,D. These successive 
naval raids by Java may be taken to indicate some control over 
the Kambuja kingdom. It is interesting to note in this connection 
a story recorded by Merchant Sulayman about the Maharaja, king 
of Zabag, an expression by which the Arab writers meant tho 
Sailendra Emperor. "It is said that once the Khmer king remark- 
ed to his minister that he would like to see the head of the king of 
Zabag before him in a dish. The Maharaja, having heard of this, 
secretly equipped one thousand vessels full of soldiers and invaded 
Khmer. The king of Khmer knew nothing of the impending danger 
until the hostile fleet had entered the river which led to his capital 
and landed its troops. The Maharaja thus took the king of Khmer 
unawares, seized upon his palace and cut off his head." 17 The 
story undoubtedly belongs to the domain of folk-lore but may have 
been based on a real struggle between Zabag and tho Khmer king- 
dom of Kambuja. It is needless for our present purpose to discuss 
whether Sanjaya, the king of Java, belonged to the Sailendra 
dynasty or not, and whether the naval raids on Champa are to be 
credited to the Sailendras or to some other kings of Java. It seems 
to be clear, however, that Java, under either the Sailendras or 
some other royal dynasty, exercised political supremacy over 
Kambuja at least for a time during the eighth century A.D. This 
sufficiently explains the dismemberment of the political fabric that 
Bhavavarman and his successors had reared in Java. The politi- 
cal association between Java and Kambuja perhaps also accounts 
for some of the striking features which we note in the subsequent 
history of Kambuja, specially the influence of Tantric religion and 
the great building activities, two features which characterised Java- 
nese culture at that time. The removal of the capital of Kambuja 
from the bank of the Mekong river to inland cities might also, not 
improbably, have been due to the fear of Javanese naval power. 
But these are all mere speculations for the present, as the history 
of Kambuja, during this period, is shrouded in darkness and no 
definite conclusion can be arrived at on these and other analogous 

17. Ibid, pp. 156-159. 



The obccurity which envelops the history of Kambuja for more 
than a century after the death of Jayavarman I is removed with 
the accession of Jayavarman II at the beginning of the ninth cen- 
tury A.D. or shortly before it. The history of Kambuja once more 
emerges into light, and we can trace her rulers in an unbroken line 
of succession down to modern times. Kambuja not only becomes 
free and united but grows in power and prestige till it becomes the 
nucleus of a mighty empire and the centre of a glorious civilisation 
whose monuments still excite the wonder of the world, This un- 
doubtedly accounts for the great honour and esteem in which the 
name of Jayavarman II was held by posterity even centuries later, 
when everything else about ancient Kambuja had faded out of 

This enthusiasm and reverence for Jayavarman II are also 
shared by modern historians of Kambuja who represent Jayavar- 
man II in brilliant limelight and depict him as a great builder and 
a powerful conqueror. Some of the most magnificent monuments 
of Kambuja have been attributed to him and he has been credited 
\vith brilliant and successful milil3ry campaigns far and wide, 
But this complacent belief has been somewhat shaken in recent 
vears by fresh facts brought to light, and it is a task of no mean 
difficulty now to write a critical account of his life and times in a 
detached spirit. But this task has to be faced, and the question 
must be treated at some length, in order tq put the history of 
Kambuja on a firm foundation. 

There is not a single record of Jayavarman II. 1 The earliest 
reference to him occurs in the inscriptions of Yasovarman who 
ascended the throne nearly half a century after the death of Jaya- 
varman II. Besides some vague and general expressions about his 
great power and suzerainty over many kings, these inscriptions 
(cf. No. 60) contain two facts of historical importance. In the first 
place the genealogical account contained in them shows that Jaya- 
varman's grandmother (mother's mother) was a niece (sister's 
daughter) of king Puskaraksa who, as mentioned above, was the 
king of Sambhupura and Aninditapura, and Jayavarman's queen 
was the niece (sister's daughter) of king Rudravarman, Secondly, 

1. The Ins of Labok Srot (No. 4, dated 703 S', has been referred to 
him by Coedes, but this is very doubtful. The point has been discussed later. 


we are told that Jayavarman II fixed his residence on the Mahen- 
dra mountain. An inscription (No. 93) of Rajendravarman, who 
flourished half a century after Yasovarman, adds only the name of 
the father of Puskaraksa, viz. king Sri Nrpatmdravarman. Half 
a century later still, an inscription of Suryavarman (No. 148) men- 
tions 724 Saka (802 A.D.) as the date of the accession of Jaya- 
varman II and gives the name of his queen as Pavitra. 2 It is not 
till we come to the reign of Udayadityavarman, about the middle of 
the eleventh century A.D., that we get the only detailed account, 
that we so far possess, about the life and reign of Jayavarman II, from 
the Sdok Kak Thorn Ins. (No. 151). In view of the importance of 
this inscription, not only from the point of view of the history of 
Jayavarman II, but also as throwing very interesting light on the 
influence of the royal priests in affairs of state, it is necessary to 
give a short account of its contents. 

It is a very long record of 340 lines which contain 130 verses 
in Sanskrit and 146 lines of prose text in the native Khmer langu- 
age. Its author was a Brahmana, Sadasiva by name, who was the 
High Priest of the Royal family and whose ancestors filled the same 
post from the time of Jayavarman II (802 A.D.) up to the year 1052 
A.D. when the record was drawn up. It gives the names of the 
kings whom they served, and we thus find here not only the royal 
names from Jayavarman II to Udayadityavarman II in an unbroken 
line of succession, but also the names of all the High Priests of the 
Tutelary Deity of Kambuja with a catalogue of the pious works 
and religious foundations of each of them, and a list of the royal 
favours in the shape of honours, dignities, grant of lands etc., 
which each received from his royal patron. Such an interesting 
history of a sacerdotal family, extending over a period of 250 years, 
is perhaps without a parallel in the history of India and her colo- 
nies abroad. This history is first recorded in Sanskrit and then 
repeated in Khmer with some variations and additional details. 

The most interesting part of this record, for our present pur- 
pose, is the account it gives of the establishment of the cult of 
Devaraja by Jayavarman II and of the first appointment of a High 
Priest of this cult with a royal decree making the office hereditary 
in his family. As the whole career of Jayavarman II is narrated 

2. Earth infers from an Inscription of Indravarman (No. 56) that Dhara- 
rundradevi was the name of a queen of Jayavarman II (Corpus, pp. 301-3). 
His arguments are not, however, convincing. 


in this connection we may quote the relevant passage in Khmer 
which runs as follows: 

"The family was dwelling in the village of Bhadrayogin in the 
district of Indrapura. Jaynvarman II came from Java to reign in 
the city of Indrapura. The venerable Guru Sivakaivalya became 
his royal priest. Then His Majesty left Indrapura, and Sivakai- 
valya accompanied him .... Having arrived at Visaya Purvadisa he 
gave lands to 6ivakaivalya and his family who followed him there. 
He also founded a village called Kuti and assigned it to them. . . . 
Then His Majesty reigned in the city of Hariharalaya. Siva- 

kaivalya also settled there with his family Then the king 

founded the city of Amarondrapura, and Sivakaivalya also settled 
there for serving His Majesty. Then the king went to reign at 
Mahendraporvata. Sivakaivalya also resided there. There His 
Majesty invited a Brahmana named Hiranyadama, versed in 
magic, in order to perform some Tantric rites so that Kambujadesa 
might no longer be dependent on Java and have a paramount ruler 
(cakravartl) of its own. This Brahmana, who came from Janapada 
(probably in India) , performed some Tantric rites (which are des- 
cribed in detail) and the worship of Devaraja. He also initiated 
Sivakaivalya into these rituals and taught him the sacred books 
dealing with them. Sivakaivalya, in his turn, taught them to all 
his relations, and the king took a vow to employ only the family 
of Sivakaivalya and none else to celebrate the worship of Devaraja. 
Then His Majesty returned to Hariharalaya and reigned there till 
his death. Sivakaivalya also died during his reign. His Majesty 
had brought Devaraja to Hariharalaya, and his successors took the 
god to various capitals which they founded in. course of time, as he 
was regarded as the protector of the realm." 3 

3. It is very difficult to foim a clear and precise idea of the cult of 
Devaraja, It seems to be the designation of the Finga (of iva) which repre- 
sented the essence of the royal authority conceived as divine and, being 
regarded as the tutelary deity, was placed in a temple on the top of a moun- 
tain, or on the summit of a pyramidal construction, representing Kailasa, the 
abode of the gods (Melanges, S. Lcm, pp. 200-202). On the other hand 
indications are not wanting that Devaraja denoted not merely, or not so much, 
a particular linga, as a ritual or ceremony, Tantrik in character. This, at any 
rate, seems to follow from the Ins. No. 151, For a full discussion on this 
point cf. BEFEO XXXIV. pp. 611-16. Bosch holds that a cult similar to 
Devaraja existed in Java (BEFEO. XXV. p. 391; TBG. LXIV, p. 227). If 
so, it is probable that Jayavarman II derived his knowledge and inspiration 
about it from that country where he resided for some time before occupying 
the throne of Kambuja. 


Leaving aside for the present the question how far we may 
rely on statements recorded two hundred and fifty years after the 
events they relate, we shall now proceed to reconstruct the history 
of Jayavarman on the basis of the data supplied by the above 

It is obvious at the very outset that Jayavarman II did not 
inherit the kingdom in a normal way. He resided for some time in 
Java, for reasons or under circumstances not known to us, and then 
returned to his native land which was under the domination of a 
foreign power ruling in Java. He freed the land from the foreign 
yoke and even went to the length of performing religious rites to 
ensure the continuity of its newly gained independence. It is pro- 
bable that he was sent by the suzerain power to rule Kambuja as a 
vassal chief, and found opportunity to proclaim his independence. 
But we have no definite information on this point, and other ex- 
planations are possible. What seems to be certain is that by some 
means or other he established an independent kingdom in 
Kambuja. 4 

That Jayavarman II did not secure the throne by right of birth 
seems to follow also from the genealogical accounts of Yasovarman 
and Rajendravarman to which reference has already been made 
above. It is true that according to the genealogy both Jayavarnion, 
and his queen were related to the royal families of Kambuja. For 
Puskaraksa, the first ancestor of Yasovarman mentioned in these 
accounts, and who ruled over both Sambhupura and Aninditapura, 
is said to have been the maternal uncle of the maternal uncle of the 
mother of king Jayavarman II. The genealogical accounts of 
Yasovarman further tell us that the mother of the queen of Jaya- 
varman II was the maternal aunt of king Prthivlndravarman who 
was the father of Indravarman and grandfather of Yasovarman. 
But the very fact that even the genealogy drawn up in the royal 
court could show no better claim to throne either for Jayavar- 
man II or his queen amounts almost to a positive evidence that he 
or his queen had no such claim worth mentioning. For nobody 
can pretend that these relationships, even if accepted as true, would 
make Jayavarman II the natural or legitimate heir to the throne. 
It is probable that they were recorded, perhaps even devised, in 

4. Maspero (pp. 28-29,) has put forward some suggestions, but they must 
be regarded as mere hypotheses, 


later times to give an appearance of legitimacy to the claim of 
Jayavarman upon the throne of Kambuja which he had actually 
seized by some means or other. That Jayavarman II did not as- 
cend the throne by right of heredity may also be concluded from 
verse 8 of the Phnom Sandak Ins. (No. 69) recorded in 817 S 
(-=895 A.D.). The royal race to which he belonged is therein 
described as the great lotus stalk which did not rise from the soil, 
and he is said to have risen like a fresh lotus for the prosperity of 
his subjects. Barth has pointed out that this evidently alluded to 
a change of dynasty, and this view appears quite reasonable. 

On the other hand a casual reference in Ins. No. 58 (v. 30) to 
Jayendradhipativarman as maternal uncle of Jayavarman II might 
indicate some legitimate claim to the throne. We do not know the 
exact status of Jayendradhipati, nor have we any idea whether he 
had any male issue, but considering the importance of daughters 
in matters of succession in Kambuja, Jayavarman II might have 
some real claim to the throne through his mother. An inscrip- 
tion (No. 50) dated 803 A.D. records the donations of queen 
Jyestharya and mentions Jayendra, queen Nrpendradevi and king 
Srmdraloka. Jayendra may be identified with Jayendradhipati- 
varman, and as the date falls early in the reign of Jayavarman II, 
it may be held that the former did not reign long before. In other 
words, Jayavarman II may be presumed to have had legitimate 
claim to the kingdom of Sambhupura, the region where this ins- 
cription was found, as the successor of Jayendradhipativarman. 

But, howsoever he might have come to the throne, the most 
important and interesting point is his frequent change of capitals. 
The location of the towns named is not free from difficulty, and we 
may briefly refer to the current views on the subject. 5 

1. Indrapura. From the data supplied by an inscription found 
at Phum Mien Coedes locates Indrapura in the district of Khbong 
Khmum in the Kompong Cham Division. He suggests that the 
actual site of the town is now represented by Bantay Prei Nokor 
whose name indicates it to be an ancient royal capital, and the 
monuments of which, although belonging to the primitive period 
of Khmer art, shows the influence of classical art in some details. 

5. Cf. discussion by Coedes (BEFEO. XXVIII. pp, 113 ff.) and P. Sterii 
(ibid. XXXVHL pp. 175 ff,) 


On the other hand P. Stern locates Indrapura at Baray near 

2. Kuti, in the Viaya (district) Purvadisa. This was proba- 
bly situated to the east of Angkor Thorn and Coedes holds, in 
common with Aymonier, that the name is still preserved in Bantay 
Kdei, though its famous temple is later in date. 6 

3. HariharSlaya.-r Aymonier's identification of this city with 
Prah Khan, immediately to the north of Angkor Thorn, had been 
generally accepted. Coedes pointed out that the inscriptions attri- 
bute a number of monuments to Indravarman who lived in Hari- 
harfilaya throughout his reign, and that these monuments are all 
to be found in a group, known as Roluos, 13 miles to the south- 
east of Angkor. He therefore provisionally located Hariharalaya 
in this region 7 and traced even a feeble echo of the last part of this 
name in modern Lolei. This theory has now been fully confirmed 
by the inscription of Kok Svay Prahm (No. 102). 

4. Amarendrapura, Here again Aymonier's identification of 
this capital with Bantay Chmar (100 miles to the north-west of 
Angkor Thorn) , and the attribution of the famous monument of 
that place to Jayavarman II, generally accepted by previous 
writers, have been challenged by M. Stern and others. Coedes 
holds the view that although the monument is later than the time 
of Jayavarman II, the site of Amarendrapura must be looked for 
in the neighbourhood, i.e., in the northern part of the province of 

5. Mahendraparvata. Aymoneir identified it with the Phnom 
Kulen (to the north-west of Angkor Thorn) , but the absence of any 
monument on the top of this hill led him to place the city of Jaya- 
varman at the foot of the hill amidst the ruins of Beng Mala. For 
the same reason Finot proposed its identification with Prah Khan. 
But the hill top contains some brick towers (Prasat Damrei Krap 
and a few other small brick buildings) intermediate in style 
between the primitive Khmer art and that of Indravarman. And 
this is all that we could expect in the reign of Jayavarman II 
according to the modern view of the evolution of Khmer art. The 

& For an account of the locality and its three ancient temples, called 
KutfSvara, cf . BEFEO. XXXVIL pp. 333 ff. 

7. This ia confirmed by the recent archaeological researches (BEFEO, 
XXXVI. p. 630)* 


location of Jayavarman's city on the top of the Phnom Kulen may, 
therefore, be accepted. 8 

If the above identifications are accepted it would follow that 
immediately after his return from Java Jayavarman fixed his 
capital at Indrapura, not far from the ancient royal seat of Sam- 
bhupura. It is noteworthy that an inscription 9 found near this 
city records the construction of gates of the temple of the Lord of 
Sambhupura by four relations of Jayavarman II. It is, therefore, 
reasonable to conclude that Jayavarman II himself was a native of 
that region and naturally set up his first capital in its neighbour- 
hood. But then we find a gradual change of royal seat towards 
the west, first towards Angkor, then further west towards Battam- 
bang, and lastly again back to Angkor. Were these changes 
merely due to royal caprices, or inspired by a desire to find a suita- 
ble site for the capital of the newly founded kingdom? It is diffi- 
cult to accept any of these views, though they have found favour 
with scholars. For all we know it may be a sign of weakness, or 
indication of troubles which forced the king to take refuge in 
different parts of the country. Considering the past history of 
Kambuja, and the almost certain fact that Jayavarman II had no 
legitimate claim to this kingdom, nothing is more natural than to 
suppose that his accession to power was not peacefully secured, 
and he had to pass many years in constant troubles which forced 
him almost to a nomadic court-life as Coedes very aptly describes 
it. In any case this is not a less reasonable hypothesis than any 
of the other two noted above. In that case our view about the life 
and reign of Jayavarman II would undergo almost a radical change. 
Instead of regarding him as a grand monarch who united the whole 
of Kambuja into a powerful kingdom, set up successive capitals 
in different parts of the kingdom, and endowed them with palaces 
and temples whose ruins lie scattered in the sites of those cities, 
we have to look upon him as an adventurer who managed to set up 
as an independent king but had to strive hard almost the whole of 
his life to secure the position he had gained against other possible 
rivals. What has been visualised as a foundation of beautiful 
capitals, one after another, may be no more than seeking refuge in 
distant corners of his kingdom against powerful foes. 

8. This, too, has been confirmed by recent Archaeological researches 
(BEFEO. XXXVI. p. 630). 

9. Aymonier I 307. 


A somewhat more favourable view is to suppose that 
the frequent change of capitals was the result of the 
chaotic political condition of Kambuja at the time of Jaya- 
varman's return from Java. Perhaps it took him many years 
to bring the whole kingdom under his control. Beginning with 
his native kingdom of Sambhupra in the east, he gradually pro- 
ceeded westwards, and the different capitals may merely indicate 
the different stages of political consolidation. Ultimately when the 
whole country was subdued, he fixed his final capital at Harihara- 
laya in the central part of the kingdom. 

A possible source of trouble for Kambuja at this period has 
generally been overlooked by scholars. The Po Nagar Inscription 
of Harivarman, 10 king of Champa, refers to one of his generals as 
having ravaged Kambuja and forcibly advanced up to the very 
heart of the kingdom. This inscription being dated in year 739 
Saka, the incident must have taken place at the beginning of the 
ninth century A.D., i.e. early in the reign of Jayavarman II. It 
is not impossible, therefore, that the Cham incursions forced Jaya- 
varman to leave Indrapura and even the Angkor region, and 
betake himself to the western part of the kingdom. It was only 
when that menace was over that he could again come back to the 
Angkor region and spend his last years in his capital Hariharalaya. 
The final choice of this capital, in place of the old Indrapura, was 
perhaps also influenced by the same consideration, viz., to remove 
the seat of the capital from the dangerous neighbourhood of the 
border of Champa. All these are possible interpretations of the 
few facts that the record of the priestly family has preserved to 
posterity and, according as we accept one or the other, we shall 
have to view the life and reign of Jayavarman II in altogether 
different lights. Thus if we hold that all the capital cities were in 
his possession at one and the same time, we must hold that he 
reigned over the whole of Kambuja, and brought about the unity 
of the country after the lapse of more than a century. But this 
has to be considerably modified if any of the other interpretations 
be accepted. 

Similar uncertainty prevails about the date of Jayavarman's 
accession. This is all the more to be regretted as until recently it 
was definitely fixed at 724 Saka (=802 A.D.) and the scholars re- 

10, Champa, Bk. Ill, p. 61; Corpiw, p, 263, 


garded it as a sheet-anchor in Kambuja chronology. This date is 
furnished by several inscriptions of Yasovarman and Suryavarman. 
But the great French scholar Coedes has drawn attention to an 
inscription in the temple of Labok Srot (No. 49) which was issued 
in the reign of king Jayavarman and is dated in the year 703 Saka 
(781 A.D.). Coedes, while first editing the inscription, shared the 
general view that Jayavarman II came to the throne in 724 Saka, 
and hence regarded king Jayavarman of this inscription as a diffe- 
rent king. But he now proposes to take this inscription as belong- 
ing to the reign of Jayavarman II and thus pushes back th? date 01 
his accession by more than twenty years. 11 Coedes reconciles this 
view with the data of the later inscriptions by interpreting the date 
724 Saka furnished by them as that of the establishment ot the 
capital of Jayavarman II on Mahendraparvata, an event which 
according to the Sdok Kok Thorn inscription (No. 151) must have 
taken place many years after his accession to the throne. 

In view of the great scholarship of M. Coedes and his un- 
rivalled knowledge of Kambuja history any hypothesis propounded 
by him commands our respect and attention, But it is difficult to 
subscribe to his present view about the date of Jayavarman II. For 
there seems to be hardly any justification for taking 724 Saka as 
the date of the capital on Mahendraparvata. It is true that the 
qualifying phrase 'Mahendr-<adri-sthiti' is often applied to Jaya- 
varman II along with the date, but it should more properly be re- 
garded as qualifying the king, and it is a too far-fetched construc- 
tion to take it along with the date. Moreover, in certain inscrip- 
tions, the passage containing the date omits all reference to 
Mahendraparvata. Thus the Prasat Kev. Ins. (No. 148) has "AsHd 
Kambuja-rajendro veda-dvi-naga-rdjya-bhak" and this may be 
compared with another verse in the same Inscription "Asld Sri 
Surya-Varmmeti veda-dvi-VLla-rajya-bhak" There can be no 
question that the latter gives the date 924 for the accession of 
Suryavarman. The identical words in the other verse should not 
be interpreted differently and we should therefore hold that Jaya- 
varman II came to the throne in 724 (802 A.D.). 12 

11. BEFEO. XXVIII. p. 119. 

12. The Prasat Kok Po Ins. (No. 123, BEFEO XXXVII. p. 389) also refers 
to the date of the accession of Jayavarman II without any reference to 
Mahendraparvata The question has been recently discussed by me in 
JG1S., X. p. 52. 


But if we are unable to accept M. Coedes' view about the date 
of Jayavarman's accession, his new theory about the date of his 
death certainly appears to be the most reasonable hypothesis. 13 
The generally accepted view that Jayavarman II died in 791 Saka 
(869 A.D.) is based on Aymonier's interpretation of the Kok Rosei 
Ins. This inscription contains a date for the accession of a king, 
but as the first part of both is missing, we only know that a certain 
king, whose name ended in Varman, ascended the throne in the 
Saka year 91 of an unknown century. Aymonier doubtfully read 
the first part of the name as Jay a, and argued that as the dates of 
accession of all the kings bearing the name Jayavarman, except 
that of Jayavarman III, are known, and none of them falls in the 
year 91 of any century, we must hold that Jayavarman III ascended 
the throne in 791 6, and consequently his father Jayavarman II 
died in that year. This view was generally accepted,but Coedes 
has very successfully demonstrated that the inscription in question 
must be referred to the reign of Jayavarman IV, and we shall 
discuss later his views about reconciling the date, which must now 
be read as 891, with the known date of his accession. In support of 
his view that the 791 Saka cannot be regarded as the date of acces- 
sion of Jayavarmap. III, Coedes has offered a new interpretation of 
the Prasat cak Ins. of Jayavarman dated the year 791 (No. 52). 
This inscription was regarded by all, including Coedes, as a defi- 
nite confirmation of Aymonier's view, and an expression contain- 
ing the words ( 16 years' was taken to refer to the age of the king. 
Coedes now interprets the expression to mean that the inscription 
was really engraved in the 16th year of the reign, and not when 
the king was 16 years old. This view seems to be quite reasona- 
ble, for while the inscriptions often refer to the regnal year it is 
very seldom, if ever, that they refer to the age of the king. Now, 
according to the new interpretation, 791 Saka was the sixteenth 
regnal year of Jayavarman III, v/ho must have, therefore, ascend- 
ed the throne in 854 A.D. This view has since been confirmed by 
the discovery of Ins. No. 51 which gives the data 782 for Jaya- 
vamian III. We may thus hold that Jayavarman II ruled from 
802 to 854 A.D. 

The names of Jayavarman's queens are known from several 
later Inscriptions. His chief queen (agra-mahist) , Pavitra by 

13. BEFEO XXVIII, pp, 113 ff, 


name, is referred to in the Prea Kev Ins. 14 Another queen, Kam~ 
bujalaksmi, called also Prana, is referred to in Phnom Prah 
Vihear Ins., which also refers to her relations as occupying high 
offices. 15 The Baku Ins. probably furnished the name of another 
queen Dharamndradevi. 16 As we have seen above, one of the 
queens of Jayavarman II was related to a royal family, and she is 
expressly called the mother of Jayavarman III (orginally known as 
Jayavardhana) , 17 Most probably she was the chief queen Pavitra. 
Jayavarman had also a son by Kambujalaksmi, Dharmavardhana 
by name. 18 An Ins. (No. 50), dated 725 S, refers to the donation 
of queen Jyestharya to Siva. As the date falls in the reign of 
Jayavarman II, she may be another queen of the same king. 

Jayavarman II revived the old tradition of Kambuja as 
against that of Fu-nan. As already noted above, Bhavavarman and 
his successors do not allude to the origin of the royal race of Kam- 
buja from Kambu and Suryavamsa, but refer instead to Kaun- 
dinya and Soma as their ancestors like the kings of Fu-nan. Jaya- 
varman's name is, however, associated with Kambu and Surya- 
vamsa. It is during his reign that an inscription of Champa 13 
refers to the country as Kambuja, whereas an earlier inscription 
of the same kingdom, dated 657 A.D., 20 refers to it as Bhavapura and 
associates it was Kaundinya and Soma. Jayavarman is referred 
to in the inscriptions of his successors as 'Kamfaujarajendra' and 
"guardian of the honour of the solar race of king Kambu," and, 
ar has just been mentioned, one of his queens bore the name or 
epithet Kambujalaksmi. After him Kambujendra and Kambu- 
jesvara became the normal official titles of the Khmer kings. Yaso- 
varman was particular in using these titles, and for a time at least 
the new capital city founded by him at Angkor Thorn was known 
as Kambupuri. Similarly the later kings of Kambuja regard them- 
selves as belonging to Siryavaiiisa and not to Somavariisa. Thus 
Jayavarman's reign marks the revival of the old legendary origin 
of Kambuja and the end of the tradition of Fu-nan. 

14. Corpus, p. 106. 

15. Ibid, p. 539, 

16. Ibid, p. 302. 

17. Ibid, p. 365. 

18. Ibid, p. 541. 

19. Champa, Bk, HI, p. 62, 

20. Ibid, p. 23. 


Like most other aspects of the career of Jayavarman II there 
is a great deal of controversy about his religion and artistic achieve- 
ments. The views held on both these subjects have to be considera- 
bly modified in the light of recent researches. 

It was believed that with Jayavarman began the golden age of 
Kambuja architecture. This belief was based mainly on the old 
identifications of the capitals of Jayavarman. For example, all 
the splendid monuments of Prah Khan were attributed to him as 
that was supposed to represent the site of Hariharalaya. Similarly 
the magnificent ruins of Beng Mala were regarded as those of his 
splendid monuments at Mahendraparvata. He was also regarded 
as having built the famous monuments at Bantay Chmar. 20a These 
views must be given up now, and even if the identification of the 
old capitals still held good, we might well doubt whether all the fine 
buildings were constructed by Jayavarman II. But this does not 
mean that he did not build any noble monuments. The probability 
rather is that he did, though we do not know much of it. Popular 
tradition ascribes to him most of the grand monuments in ancient 
Kambuja. These are believed to have been erected by him with 
the help of an architect who was the son of a nymph and learned 
architecture from the gods in heaven. No other Kambuja king 
has left such a deep impress upon posterity as a great builder, 
and there must have been some basis on which the popular legend 
has grown up. It is a pity that our knowledge of his artistic activities 
is so meagre and uncertain. But we may well believe that he 
made a distinct contribution to the development of Kambuja 
architecture which reached such a high pitch of grandeur and ex- 
cellence under his successors. 

The view that Jayavarman II was a Buddhist similarly rests 
upon very weak foundations. This conclusion is based mainly on 
two grounds viz., that Jayavarman came from Java, which was a 
centre of Buddhism, and he founded Amarendrapura, now repre- 
sented by Bantay Chmar, whose sculptures are predominantly 
Buddhist. 21 None of these proves much, and the identification in 
the latter case is open to doubt as mentioned above. 

20a. Finot even suggested that Jayavarman began the construction of the 
great capital which bore later A*\ the name of Yasodharapura (IHQ. I, p. 615|>. 
21, 1HQ, I, p. 616, 


The only positive evidence regarding the religious beliefs of 
Jayavarman II is furnished by the Sdok Kak Thorn Inscription. 
(No. 151). Sivakaivalya, the royal priest, and the priestly family 
founded by him, which supplied royal chaplain for two hundred 
and fifty years, were undoubtedly gaiva and presumalJly the king 
followed the same religion. The cult which he established as state 
religion with the help of the Brahmana Hiranyadama seems to be a 
form of Tantric Saivism. This follows from the detailed descrip- 
tion of the magic rites contained in the Sdok Kak Thorn Ins. It 
is said that the Brahmana Hiranyadama performed the ritual as 
laid down in Vinasikha and consecrated the Devaraja cult He 
taught Sivakaivalya the four texts known as Vinasikha, Nayottara, 
Sammoha and Sirascheda. He recited these texts from beginning 
to end, so that they may be put in writing for the use of Sivakai- 
valya, and he taught the latter how to conduct the ritual of Deva- 
raja. Later, \vo are told that these four texts constituted the four 
faces of Tumburu, 

Now Dr, P. C. Bagchi has shown 22 that one of the four texts, 
viz. Nayottara, is definitely known to belong to the Agama proper, 
i.e., the oldest Saivite canon (which conformed to the Vedas and 
had not entirely separated from the Vedic religion like the later 
Saiva sects), and the other three texts belong to the Saiva canon 
which grew later under its inspiration, These four Tantric texts 
were authentic Saiva-sastras studied in India in the 7th and 8th 
centuries A.D. if not earlier, and Tumburu is definitely described 
in Yoga-Vasistha Ramayana as an aspect of Rudra. These texts 
were introduced in Kambuja for establishing the rites known as 
Devaraja, which, therefore, must represent a 6aiva cult. In the 
Sanskrit text of the inscription the cult is referred to as 'Siddhi* 
called Devaraja, but it appears from other passages that Devaraja 
was a phallic representation of Siva. 23 Thus the state-religion 
established in Kambuja by king Jayavarman II was a form of 
Tantrik Saivism, which included some mystic rites and was based 
on the four Saivasastras specified by name. That the king him- 
self was a follower of the same religion can hardly be doubted. Ho 
is said in Ins. No. 89 to have performed ten millions (koti) of 

22. 1HQ. VI, p. 97. 

23. See ante, f.n. (3) above. 


Jayavarman's decision that the royal priest should be selec- 
ted from the family of 6ivakaivalya alone was also probably due 
to the strict adherence to the Saiva aga>nas according to which the 
Sivacaryas had to be chosen preferably from the Brahmanical 
families of North Indian origin. Such families with knowledge of 
Agama-sastras were probably rare in Kambuja and hence the 
choice was confined to a single family. Parallel instances may be 
quoted from India. The great Cola king Rajondra Cola is said to 
have appointed Sarvasiva Pandita Sivacarya as the priest of the 
Rajarajesvara temple at Tanjore, and ordered that in future his 
slsyas and their sisyas alone belonging to the Aryadesa, the 
Madhyadesa and the Gauda-desa shall be eligible for the office of 
chief priest. The Malla kings of Bhatgaon (Nepal) also had Brah- 
mins from Bengal as their priests. 24 In any case there is hardly 
any doubt that the Kambuja Court was strongly influenced by 
Saivism then prevailing in India, 

Although there are few specific facts about Jayavarman II 
which can be said to be established beyond dispute, there is no 
doubt thai Jaynvrmnan II played an important rolo in the history 
of Kambuja. He delivered it from foreign yoke, first of Java and 
then of Champa, and gave the kingdom a unity and solidarity 
which it had lacked for a century. The Devaraja cult introduced 
by him remained the official religion for a long period. His final 
choice of the capital at Hariharalaya was destined to make the 
region of Angkor famous in the world on account of many grand 
palaces and temples built there by successive kings. All these 
explain the eminent position which Jayavarman occupied for cen- 
turies in the history of Kambuja, a fact testified to by most flatter- 
ing references to him in a large number of later inscriptions. 
Most fulsome praise, for example, is bstowed on Jayavarman II 
in the Phnom Sandak Ins. (No. 69) of the time of Yasovarman. After 
reference to his beauty and glory in the most extravagant phrases 
it adds a verse which has a double sense applicable both to the 
king and the grammarian Panini, and extols the former's know- 
ledge of the great grammar. Although such phrases generally do 
not mean much, their occurrence in an inscription long after his 
death undoubtedly indicates his power and popularity. Even now 
he is the divine hero of Kambuja which represents him as the son 
of Indra. The sacred sword of Kambuja, which is still used by the 

24. IHQ, VI, p. 100, 


Kambuja kings at the time of their coronation, and jealously guard- 
ed by priests who claim descent from the old Brahmanas, is suppos- 
ed to have been a relic of Jayavarman II who remains the national 
hero and a great landmark in Kambuja history. 25 

Jayavarman II died, as noted before, in 854 A.D. and re- 
ceived, after death, the name of Paramesvara. This kind of 
posthumous name was usually formed by adding the word 'loka' 
or 'pada' to a divine name (Brohma, Visn.u, Siva, Indra etc.). 
Almost all the successors of Jayavarman II possessed such names, 
but with one or two exceptions, such name is not associated with 
any predecessor of Jayavarman II. 

After the death of Jayavarman II his son Jayavardhana ascended 
the throne under the name of Jayavarman (III). The inscription 
of Prasat Cak (No. 53) in the region of Angkor, which refers to the 
year 791 S (^801) A,D.) as the sixteenth year of his reign, mentions 
in detail a story of his hunting elephants. Other inscriptions also 
refer to his capturing elephants and one of them has even preserv- 
ed the name of the chief of the royal elephant hunters. Except this 
inordinate passion for elephant hunting we do not know anything 
about him. He ruled from 854 to c, 877 A.D., and his posthumous 
name was Visnuloka. With him ends the line of Jayavarman II. 

According to the Chinese chronicle Man-Chu written in 863 
A.D. the Khmer Empire extended in the north to Chen-nan of Nan- 
chao which most probably corresponded to the northern part of 
Alavirastra, to the west of Tonkin. 26 As the author gathered his 
information by a personal visit to these regions in 862 A.D., we 
may regard it as true of the period of Jayavarman III. It would, 
therefore, follow that his kingdom included the whole of Laos in the 
north and almost touched the frontier of Yunnan. Of course it is 
impossible to say to what extent this is due to his own conquests. 
For it is likely that Jayavarman II not only united the whole of 
Kambuja but also added Laos to his dominions, and this would be 
quite in keeping with the traditional glory of Jayavarman II to 
which reference has been made above. In any case it is tolerably 
certain that under Jayavarman II or his son the Kambuja kingdom 
had developed into a powerful empire. 

25. Maspero, p. 31. 

26. Et. AS. II, p. 94. 



This is corroborated by the Arab writers. Ya'kubl, writing 
about 875 or 880 A.D., describes the Khmer kingdom as vast and 
powerful, the king of which receives homage of other kings. 27 
Another Arab writer, Ibn Rosteh (903 A.D.), refers to the high 
standard of administration in the Khmer country. "There are", he 
says, "eighty judges. Even if a son of the king appears before then* 
they would judge equitably and treat him as an ordinary complai- 
nant*" We are further told by the same writer that "the principal 
revenue is derived from cock-fight which brings the king fifty mans 
of gold per day". 28 Masudi, who wrote in 943, but evidently got his 
information from older writers as he repeats a great deal of their 
accounts, adds that the Khmer troops consist mainly of infantry 
because their country is full of hills and valleys, rather than plains 
and plateaus. 29 

Several Arab writers bestow high praises on the Khmers for 
their abstinence from wine and adultery (debauchery). Thus Ibn 
Khordadzbeh (844-848 A.D.) says: "The kings and peoples of India 
abstain from drinking of wine but they do not consider adultery 
is an illicit act, with the sole exception of the Khmer king who 
forbids both drinking and adultery". 30 This is repeated by Ibn 
Rosteh (903 A.D.) on the authority of an Arab traveller Abdullah 
Muhammad bin Ishak who lived in the Khmer country for two 
years. "During this period," says he, "I have never seen a king 
more opposed to sexual license and more severe against drinking, 
for he inflicts capital punishment for both the offences." 31 The 
same view is recorded by Abu Zayd (c. 916 A.D.). 32 But the kings 
had large harems. Ibn Al Faklh (902 A.D.) says that the king 
maintains four thousand concubines, 33 

27. Ferrand-Te-rtes, I. p. 48. 

28. Ibid, pp. 71, 78. 

29. Ibid, p. 93. The passage is, however, translated differently by Dr, S. 
M. H. Nainar in Arab Geographers' knowledge of Southern India (Univ, of 
Madras, 1942). According to him "the inhabitants mostly go on foot because 
their country is full of mountains and valleys, few plains and table lands'* 
(p. 174). 

30. Ferrand Textes, I., p. 28. 

31. Ibid, pp. 69-70. 

32. Ibid, p. 85. 

33. /bid, p, 64. 



With the death of Jayavarman III ended the direct line of Jaya- 
varman II, and one Indravarman ascended the throne in 799 S' 
(=877 A.D.) (Nos. 54, 89). He was very remotely related to the 
queen of Jayavarman II. For we learn from several inscriptions 
(Nos. 56, 60-2) that his father king Sri Prthivlndravarman was the 
son of the maternal aunt of the queen of Jayavarman II, and his 
mother was the daughter of king Sri Rudravarman and daughter's 
daughter of king Sri-Nrpatmdravarman. King Rudravarman was 
also the maternal uncle of the queen of Jayavarman II. These 
relationships will be clear from the genealogical table given 

X Nrpatmdravarman 

i i r i 

D D Rudravarman=rD 

I t I 

Jayavarman II~D Prthivlndravarman- D 

I ! 

Jayavarman III Indravarman 

Where and when kings Nrpallndravarman, Rudravarman and 
Prthivlndravarman ruled it is difficult to say. They were either 
local chiefs ruling before Jayavarman II, or were vassals of the 
latter. In any case nothing is known of the reign of any of these 
three. It is difficult, therefore, to judge of the right of Indravarman 
to the throne of Kambuja, and we are ignorant of the circum- 
stances under which he came to the throne. 

It has been suggested that the marriage of Indravarman might 
hove paved the way for his accession to the throne. According to 
the genealogical account of Yasovarman (No, 60), Indradevi, the 
queen of Indravarman, was the daughter of king Mahipativarman, 
and this Mahipativarman was the son of Rajendravarman and his 
queen NrpatindradevL The same inscription informs us that 
Rajendravarman was connected with the royal family of Vyadha- 
pur a through his mother, and was a descendant of Puskaraksa, who 
had united the kingdoms of Sambhupura and Aninditapura under 
his rule. The mother of Indradevi, named Rajendradevl, was 
descended from a royal family founded by Agastya a Brahmana 



from Aryadesa (i.e., India). These relationships, as well as the 
connection of Jayavarman II with these families, already explained 
above, will be clear from the following genealogical table 

Agastya (a = 

Brahmana from j 

India) | 


Yasomati (a 
royal princess) 

Ruler of 

Nar endr alaksm I = Raj apati varraan 


(Ruler of 

Puskaraksa D 
(also ruler of | 
ambhupura) D 


D - S D 

I I 

Rajendiavarman Jayavarman II 
( Nrpatlndradevi) (802-854 A.D.) 

I I 

MahTpativarman Jayavarman III 

1 (854-870 A.D.) 


(ace. 877 A.D.) 

This genealogy also does not give any clear or uncontestable right 
of succession to Indravarman through his queen, and raises the 
same doubt about the position of king Mahipativarman and his 
predecessors vis a vis Jayavarman II and III. Here, again, we are 
forced to conclude that the royal ancestors of Indradevi were either 
local chiefs ruling before Jayavarman II or vassals of the latter 
and his son. 

On the whole the genealogies of Indravarman and his queen 
seem to indicate that in addition to the two kingdoms of Sambhu- 
pura and Aninditapura there were other local kingdoms in Kam- 
buja in the eighth century A.D., some of which probably continued 
as vassal states even during the reigns of Jayavarman II and 
Jayavarman III. 

It is reasonable to conclude that Indravarman originally be- 
longe'd to one of these states, and either the absence of any legiti- 
mate heir of Jayavarman III or some other circumstances, of which 
we have no knowledge, enabled him to secure the throne. It may 
be assumed that he did not rebel against the family of Jayavarman 
and come to the throne by violent means. For his inscriptions and 
those of his successors refer to Jayavarman II and III with respect 
and he appointed as his guru the grandson of the maternal uncle of 
Jayavarman II (No. 58) . 

It may be concluded from the epigraphic records that the whole 
of Cambodia had by this time been consolidated into a happy, rich 


arid prosperous kingdom. Indravarman claims in his record (No, 
58) that his commands were respectfully obeyed by the rulers of 
Cma, Champa and Yavadvipa. Such specific claims are not usually 
met with in the inscriptions of the Kambuja rulers and cannot b'> 
ignored as mere bombasts or figments of imagination. As regards 
Champa we have already noted that one of its generals advanced 
up to the heart of Kambuja and ravaged the kingdom early in the 
ninth century A,D. We may, therefore, presume that the struggle 
between the two kingdoms continued practically throughout the 
ninth century A.D. Indravarman's contemporary on the throne of 
Champa was a king bearing the same name who probably founded 
a new dynasty. We do not learn anything form the history of 
Champa which would either prove or disprove the claim of the 
Kambuja king. But we find about this time references to diplomatic 
missions from Champa to Java, 1 This latter must be the kingdom 
of Yavadvipa mentioned by the Kambuja ruler as his vassal state. 
The diplomatic alliance between two states over both of which 
the Kambuja ruler claims supremacy may not be without signifi- 
cance. The reign of Indravarman coincides with an obscure period 
in Javanese history which saw the end of the kingdom of Mataram 
in Central Java and the shifting of the centre of political authority 
and Indo-Javancse culture to the eastern part of the island. 2 To 
what extent, if any, this final abandonment of Central Java was 
due to the rising power of Kambnja, we cannot say. But it is not 
unlikely that Kambuja, which suffered at the hands of both Champa 
and Java towards the close of the eighth and the beginning of the 
ninth century A.D., now turned against her old enemies and 
obtained some success. Unfortunately we have no definite know- 
ledge of these events, 

Indravarman's claim to supremacy over China is more 
puzzling, on the face of it, and will be discussed later. 

Indravarman was a great builder. Ins. No. 54 (v. 7) informs 
us that immediately after ascending the throne he made a pro- 
mise (pratijndm krtavdri) that within five days counting from 
that very day, he would begin the work of construction 
(prdrapsye khananddikam) . The next verse tells us that he had 
constructed, according to his own design, a simMsana (royal 
throne), the vehicle called Indrayana, Indravimanaka, and Indra- 
prasadaka (probably two palaces), all made of gold (haima) . He 

1. Champa, p. 62. 

2. Suvarnadvipa, I, pp. 237 ff. 


installed three images of Siva and three of the goddess (Durga), 
which were works of his own art (svo-silpo-racita, v. 28). His 
various religious endowments, including temples and images of 
gods and a big tank called Indratataka, are referred to in other 
inscriptions. His reign marks an important stage in the deve- 
lopment of Kambuja art. Parmentier has made a special study 
of the monuments that may be definitely ascribed to Indravar- 
man, and in his opinion, the art of Indravarman forms an inter- 
mediate stage between the Primitive and Classical art of Kambuja. 

Indravarman ruled for only twelve years (877-889) and 
received the posthumous title Isvaraloka. He was succeeded by 
his son Yasovardhana under the name Yasovarman. Yasovarman 
occupies a place of honour in the history of Kambuja and his name 
has been immortalised by the foundation of a new capital city, on 
the top of the hill called Phnom Bakhen, which was at first called 
Kambupurl and later Yasodharapura. Although this is not the 
famous city of Angkor Thorn, covered with magnificent ruins, as 
was firmly believed until recent years, it extended round the hill 
and included a large part of the present site of Angkor Thorn, and 
Yasovarman may still be credited as the founder of Angkor, though 
in a qualified sense. The region round his newly founded capital 
city remained the heart of Kambuja power and culture till the last 
day of its greatness. He may be also said to have laid the foun- 
dation of the Angkor civilisation whose glory and splendour form 
the most brilliant chapter in the history of Kambuja. 

Indravarman placed Vamasiva, the grand-nephew of Sivakai- 
valya, in charge of the education of Yasovarman (No. 151 D. vv. 
4-10) . Yasovarman is said to have been fond of 6astras and Kavyas 
(62 E.D. 1), and a perusal of his inscriptions leaves no doubt that 
Sanskrit literature, both secular and religious, was highly patronised 
in his court. These inscriptions are, however, poor in historical 
material. Reference is made to the numerous military campaigns 
of the king, including a naval expedition (62 D-B. 19) , and he is 
said to have reinstated vanquished kings (62C-C. 5) and married 
their daughters (62C-B. 27). But they do not refer to specific 
events of his reign. The dominions over which Yasovarman ruled 
were extensive. On the north it reached the frontiers of China, 
and on the west, the mountains which form the watershed between 
the rivers Menam and Salween. The eastern and southern boun- 
daries were formed respectively by the kingdom of Champa and 
the sea. 


King Yasovarman has left numerous records (Nos. 60-73), 
Some of these refer to religious endowments and construction of 
sanctuaries while others give detailed regulations of the large 
number of monasteries founded by him. 

The inscriptions of Yasovarman are distinguished by two pecu- 
liarities. In the first place one single record (No. 60) is reproduced, 
in identical words, no less than eleven times. Secondly, the texts 
of these eleven inscriptions and of another (No. 61) are written 
twice, once in the ordinary alphabet used in Kambuja in those 
days, and again in a novel type of alphabet which has a close resem- 
blance to the North Indian scripts. Further, seven inscriptions are 
written in this latter alphabet alone. 2a 

These inscriptions of Yasovarman, in spite of vague generali- 
ties which they contain, give us many interesting sidelights on the 
various phases of his internal administration. They hold out a 
picture of a happy, prosperous and peaceful kingdom ruled over 
by an able and wise monarch who took all possible measures to 
ensure the welfare of the kingdom in all its aspects, political, 
economic, religious and social. The elaborate regulations framed 
by him give us an insight into the social and religious condition 
of the time and the earnest effort made by the king to improve it. 
Making all due allowances for exaggerations of court poets we 
must regard Yasovarman as a brave general and ideal king, shining 
equally well in arts of war and peace. Himself a great scholar, 
he was a patron of art and science. He was liberal in his religious 
views, and although a devoted follower of Saivism, he patronised 
Buddhism in an unstinted manner. He was a great king in every 
sense of the term. Perhaps the court poet did not exaggerate very 
much when he said that the glory of Yasovarman was sung even 
after his death, by the people "in their games, on their beds, and 
in their travels" (No. 70). Yasovarman received the very appro- 
priate posthumous title of Paramasivaloka. 

Yasovarman died about 908 A.D. The history of Kambuja 
during the next twenty years is somewhat obscure and uncertain. 
We know that two sons of Yasovarman, viz. Har^avarman I and 
lanavarman II ascended the throne one after another, and next 
came Jayavarman IV, the husband of the sister of Yasovarman. 
But the known dates of these kings cannot be easily reconciled with 

2a. Nos, 62 <A-E), 63A, 67, 


a normal course of succession. The Sanskrit text of the Ins. No. 75 
refers to a donation by king Harsavarman, and the fragmentary 
Khmer text of the inscription contains the date 834 Saka (912 
A.D.), but what connection, if any, this date has with the donation 
is not clear, 3 and we cannot therefore, say whether it falls during 
the reign of Harsavarman, The Ins. No. 74 refers to Yasovarman 
and his two sons and mentions a foundation made in 832 Saka 
(=910 A.D.), but there is nothing in the context to show that this 
date falls within the reign of the last-named king viz, Isanavar*nan 
II. The Ins. No. 78 is dated in 844 aka (=922 A.D.), in the 
reign of a king whose name begins with Ha, but as the next letter 
is not very legible, its restoration as Harsavarman, though very 
probable, is by no means certain. The only konwn date of the 
sons of Yasovarman is furnished by Ins. No. 98 which refers to an 
address (nivedana) presented to king Isanavarman II in 847 S' 
(=^925 A.D.). This certainly proves that Harsavarman must have 
ceased to rule, and his younger brother Isanavarman II ascended 
the throne some time before 925 A.D. 

It is, however, difficult to reconcile this with the known dates 
of Jayavarman IV. Although Ins. No. 80 gives 928 A.D. as the date 
of his accession, we have actually one inscription (No. 76) dated 
in 921 A.D., belonging to his reign. What is more important, this 
inscription as well as two others (Nos, 77-77A), dated 921 A.D., 
found at Koh Ker, show that the tutelary deity of the Kambuja 
royal family was already transferred to that place, and conse- 
quently the capital city must have been removed there, This goes 
against the view, formerly held, that Jayavarman ruled as Viceroy 
at Koh Ker during the rule of Yasovarman's son or sons, Finally 
the relationship between Jayavarman IV and his predecessors is 
not such as would induce us readily to believe that he had legiti- 
mate claim to the throne, and an expression in Ins, No. 84 (v. 2) 
implies that he came to the throne by his own might and not by 
any right of succession. 

The most reasonable inference from the above facts seems to 
be that Jayavarman rebelled against Isanavarman II and set up as 
an independent king even during his life-time, some time before 
921 A.D., though in a later age, an attempt was made to obliterate 
the memory of the unpleasant incident by regarding his formal 

3. Cf. Corpus, p. 552, JL 1. 


accession to the throne to have taken place in 928 A.D., probably 
the year in which Isanavarman II died. 4 

Nothing more is known of Harsavarman and Isanavarman II, 
beyond the fact that their posthumous names were respectively 
Rudraloka and Parama-Rudraloka. 

The most important event of the reign of Jayavarman IV was 
the removal of the capital to Koh Ker (Chok Gargyar) (No. 151) . 
Evidently the new ruler did not like, or think it prudent, to remain 
at a city which was so intimately associated with the rulers whose 
throne he had usurped. 

The choice of the new capital cannot be regarded as a happy 
one. It was situated in a wild barren country about 50 miles north- 
east of Angkor. Among the extant ruins we can still trace the 
remains of one principal and twelve subsidiary temples, and the 
usual artificial lake. A characteristic feature was a temple con- 
taining three huge monolithic lingas fashioned out of three natural 
boulders. This perhaps accounts for the fact that the town was 
not oriented as usual, its north-south axis being considerably in- 
clined towards the west. Jayavarman also erected a pyramid more 
than 40 yards high, which was designed as the pedestal of the tute- 
lary deity Devaraja. 5 

Jayavarman IV is described as having destroyed the ruler of 
Champa (No. 83A) . This probably implies renewed hostilities with 
the eternal enemy on the eastern border, but we possess no detailed 
account of the struggle. 

Jayavarman had the posthumous name of Paramasivapada and 
was succeeded by his son Hansavarman II: According to the 
generally accepted reading of the Ins. No. 85 the king ascended the 
throne in 864 S' (942 A.D.). But the Ins. No. 84 belonging to 
his reign appears to be dated in 863 S' (=941 A.D.). As the other 
inscription is now lost and its reading cannot be checked, we can- 
not altogether dismiss the data supplied by the Ins. No. 84. The 
date of the accession of Harsavarman II may, therefore, be earlier 
by at least one year than that generally accepted. 6 His posthumous 
name was Brahmaloka. 

4. Of. BEFEO. XXXI, pp. 12 ft. 

5. Inscriptions, pp. 69-70* 

6. /bid, p. 261. 



Harsavarman II was succeeded by Rajendravarman. Several 
inscriptions (Nos. 88, 89) refer to Rajendravarman as the elder 
brother of Harjsavarman II, and hence it has been generally 
held that Rajendravarman was the elder son of Jayavarman IV. 
Why the elder son reigned after the younger remained a mystery. 
This mystery was still more deepened by the Mebon Ins. (No. 89A) 
which says that Rajendravarman was the son of Mahendravarman 
and Mahendradevi. According to this inscription both Mahendra- 
varman and his father ruled over a kingdom (the name of which 
except the last two syllables 'pura' has disappeared), and 
Mahendradevi was descended on the mother's side from Baladitya, 
king of Aninditapura. Finot, who edited this inscription, tried to 
reconcile its data with the older view by suggesting that Mahendra- 
varman was but another name of Jayavarman IV. Coedes however 
very properly rejected this hypothesis and offered an alternative 
! view that Rajendravarman was the son of Mahendradevi by a pre- 
vious husband Mahendravarman. All doubts have, however, been 
set at rest by the discovery of the Pre Rup Ins. (No. 93). We 
learn from it that Yasovarman had two sisters Jayadevi and 
Mahendradevi. Jayadevi was married to Jayavarman IV and their 
son was Harsavarman II. Mahendradevi was married to Mahendra- 
varman and their son was Rajendravarman. Presumably Jayadevi 
was the elder sister, and hence her son Harsavarman, although 
younger in age than Rajendravarman, had the prior right to 
succession. The Pre Rup Ins. further mentions Vedavati, a 
descendant, by way of female line, of Sarasvati, sister's daughter of 
Baladitya who was descended from Kaundinya and Soma, We 
are told that Mahendravarman was descended from the royal family 
of the father of Vedavati while Mahendradevi was a descendant of 
Vedavati herself. The genealogy may be explained by the follow- 
ing Table: 

Kau ndiny a =: Soma 

Baladitya D 



Vedavati ~D viveda 


The succession after Yasovarman may be shown by the 
following table: 

I. Indravarman 

I I , I 

V. Jayavarman XVJayadevi II. Yasovarman Mahendradevi 

| | Mahendravarman 

VI. Harsa- |" _i | 

varman IL III. Harsa- IV. Isana- VII. Rajendra- 

varman I. varman II varman 

The succession of kings, marked by Roman numerals IV to VII, 
can only be regarded as natural and legitimate by supposing that 
Nos. Ill, IV, and VI left no issue at the time of their death. This 
is too unusual to be readily believed, and it is more likely that the 
whole period was marked by a struggle for succession among 
persons who were in some way connected with the royal family. 
We have already cited evidence which seems to corroborate this, 
at least in regard to Jayavarman IV. 

Rajendravarman ascended the throne in 866 S' (rr944 A.D.) 
at an early age (No. 93, vv. 27, 37, 52) . The most important event 
in his reign is the removal of the capital back again to Yasodhara- 
giri or Yasodharapura, the city on the top of the Phnon Bakheu 
hill, founded by Yasovarman. The Ins. No. 92 informs us that 
the king embellished the city which was deserted for a long time. 
There are some hints in the Ins. No. 93 that he had seized the royal 
power after some contest. 7 According to Aymonier the old Khmer 
story of prince Baksei (Skt. paksi bird) Chan Krang refers to 
Rajendravarman. 8 That would, in effect, mean that his brother, 
the reigning king, was disturbed by a prophecy that the latter 
would be dethroned by him. Rajendravarman consequently had 
to fly for his life and ultimately defeated and killed his brother. 
Such a fratricidal struggle is not of course an improbable one, and 
that would at least satisfactorily explain the desertion of the capital 
associated with the family of his rival for two generations. 

We possess a large number of very long prasastis of the reign 
of Rajendravarman but they do not throw much light on the politi- 
cal history of the period. Several inscriptions (Nos. 92, 93, 97) 
mention that he defeated the hostile kingdoms, particularly that of 

7. Ibid, p. 75. 
8. Aymonier I, p. 219. 


Rajendravarman undoubtedly gained some success in his 
expedition against Champa. An Ins. from Champa informs us 
that the Kambujas had carried away the golden image of the Po 
Nagar Temple and the king of Champa had installed in its place 
a stone image in the year 887 (965 A.D). Another Ins. in the 
same temple records the installation of the golden image of the 
goddess Bhagavati in 840 (=918 A.D.), and the Kambuja invasion 
must have therefore taken place between these two dates. As the 
stone image was installed in 965 A.D., and such an important temple 
as that of Po Nagar was not likely to be left empty for a long 
time, the Kambuja invasion mentioned in the Po Nagar inscription 
probably refers to that of Rajendravarman. It would show that the 
Kambuja army advanced up to Khan Hoa province, and severely 
defeated the Chams. 

Rajendravarman is also credited in his inscription (No. 93) 
with victorious campaigns in all directions, north, soulh, east and 
west, but no details are given, though as we shall see later, these 
may not be mere empty boasts. 

Rajendravarman received the posthumous name of Sivaloka. 
He probably appointed his son Jayavamnan as his regent in 968 
A,D. and next year, on his death, the latter obtained full 
sovereignty. 9 

The reign of Jayavarman V was marked by a predominance of 
Buddhism as is clearly evidenced by the inscription (No. 106) 
engraved by his minister Klrtipandita. It contains the instructions 
and regulations issued by the king for the propagation of Buddhist 
doctrines. But Saivism remained the official religion. 

Jayavarman also erected some notable monuments, such as 
Hemasrnga-giri but its identification is uncertain. 10 He continued 
the aggressive policy against Champa and obtained some success 
(No. 114). He died in 1001 A.D. and his posthumous name was 

The period of a century and a quarter (877-1001 A.D.) covered 
by the reigns of Indravarman and his seven successors constitutes 
a landmark in the history of Kambuja, and it is necessary to empha- 
sise some of the special features which characterise it. 

9. BEFEO. XXVIII. p. 115, 
10, Ibid, pp. 82-3, 


In the first place, the period witnessed a great extension of the 
political power of Kambuja and therewith also the sphere of Indian 
culture and civilisation. This is proved by the Chinese annals 
which have fortunately preserved for us a fairly comprehensive 
picture of the political geography of Indo-China about the year 
960 A.D. when the Song Dynasty began to rule in China. This, 
together with the information supplied by the chronicles of Burma 
and Slam, enables us to make a broad survey of the political condi- 
tion of the whole of Indo-Chinese Peninsula lying to the south- 
east of India and south-west of China about the middle of the tenth 
century A.D. 11 

In the year 960 A.D. when Chao Kuang-in, the founder of the 
Song dynasty, seized the government of China, the whole of Tonkin 
with the districts of Than-hoa and Nghe-an lying to the south of 
it constituted the Chinese province of Ngan-nan. But the Chinese 
authority over this province was only nominal, for since the begin- 
ning of the tenth century A.D. it practically functioned as an 
independent state. In 968 A.D. the Chinese formally acknowledged 
its independence when its ruler Dinh Bo-linh proclaimed himself 
to be an Emperor and gave the new name Dai-co-viet to his 

To the north and north-west of this kingdom was the indepen- 
dent Hinduised Thai principality of Nan Chao (in N. Yunnan) 
that had freed itself from the Chinese yoke about 730 A.D. The 
Tang dynasty maintained a protracted struggle to re-establish its 
authority over this region, but failed. Indeed so painful was the 
memory of this fruitless campaign to the Chinese that when the 
general of the first Song Emperor proposed to his master to make 
another attempt to reconquer Nan Chao, the latter, reflecting upon 
the disasters sustained by the Chinese under the Tang dynasty, 
refused to have anything to do with it, and for the next three 
centuries Nan Chao remained an independent principality. With 
the loss of these two southern provinces the Chinese Emperor lost 
direct contact with the Indo-Chinese Peninsula. 

To the south and west of these lay the three well-defined 
Hinduised kingdoms viz, Champa in the east, corresponding to 
Annam, Ramannadesa in the west corresponding to lower Burma, 
and Kambuja in the south. The central region of the peninsula 

11. For details, cf. Et. A?. II. pp. 79 ft. 


surrounded by these kingdoms was peopled by the Thais who had 
imbibed rudiments of Hindu civilisation. They set up a number 
of principalities which bore Hindu or Hinduised names, but it 
is not always easy to locate these kingdoms definitely. 

The Kambuja inscriptions of this period, as we have seen 
above, refer to the conquest of China by some of its kings. The 
references are vague and general and, by themselves, do not enable 
us to form any definite idea of the nature and extent of the Kam- 
buja power in respect of that mighty oriental kingdom. But it is 
evident from the Chinese annals that although Kambuja had noth- 
ing to do with the territory of China proper, it established its 
suzerainty over many states in the Indo-Chinese Peninsula which 
were formerly within the zone of the political supremacy of China, 
and this geographical expression in the Kambuja records must be 
construed accordingly. 

As already noted above, the extension of Kambuja power in 
these northern regions can be traced as early as 863 A.D. But 
during the period under review the Kambuja sovereignty and 
Hindu culture seem to have been definitely established throughout 
the region up to the borders of the modern Chinese province Yun- 
nan. The kingdom, which the Chinese call Nan-chao and is referred 
to as Mithilarastra in Thai chronicles, comprised the northern 
part of Yunnan. Immediately to its south lay the kingdom which 
is called Alavirastra, the kingdom of the giant Alavi, It comprised 
the southern part of Yunnan. According to a contemporary Chinese 
chronicler, who visited these regions in 862 A.D., the northern part 
of Alavirastra formed the boundary of the Khmer empire. When, 
therefore, Indravarman claims that his commands were obeyed by 
the king of China, and Yasovarman asserts that his empire reached 
up to the frontier of China, we must presume a further expansion 
of the power of Kambuja at the cost of Mithilarastra (Chinese Nan- 
chao), which would extend the Kambuja power into the heart of 
Yunnan, probably not far from the border of the then kingdom of 
China. The memory of this Kambuja empire is preserved in the 
local annals. The chronicles of Yonaka which compirsed the two 
kingdoms of Alavirastra and Haribhunjaya record the foundation of 
Suvarnagrama, the site of the later capital Xien Sen, by a Khmer 
emperor. The chronicle of Bayao, a town about 60 miles further 
south, on a branch of the upper Mekong river, states that ruins of 
old palaces and cities belonging to the old time of Khmer kings 
were scattered in mountains and forests when this city was founded. 
The victorious campaigns of Rajendravarman in all directions 


evidently relate to his campaigns in these regions. On the whole 
it may be safely presumed that throughout the reign of Indravar- 
man's dynasty the Kambuja Empire\extended in the north as far 
as Yunnan and included a considerable portion of it. 

While the Kambuja kingdom was thus expanding along the 
valley of the Mekong river towards the north, it also extended its 
authority along the valley of the Menam on the west. In this 
region, which now constitutes the home provinces of the kingdom 
of Siam or Thailand, the country of Lavapuri, comprising all the 
territory between the Gulf of Siam in the south ami Kampheng 
Phet on the north, formed a stronghold of Kambuja power. For a 
long time this was regarded as an integral part of the Kambuja 
kingdom. But the Kambuja kings also exercised political influence 
over the petty principalities of the local ruling chiefs that lay to 
its north. The successive kingdoms in this region in geographical 
order beginning from the south are Sukhodaya, Yonakarastra and 
Ksmerarastra which touched the Kambuja kingdom of Alavirastra 
on the Mekong valley. The chronicles of these kingdoms refer to 
the Kambuja sovereignty over them and the very name Ksmera- 
rastra of the northernmost of these recalls the suzerainty of that 
people throughout the Menam valley. The Kambuja kings esta- 
blished a strongly fortified post at a place called Unmargasilanagara 
which commanded the roads to the upper valleys of both the 
Mekong and the Menam rivers, and although the petty vassal states 
on the Menam often revolted against the Kambuja authority, the 
Kambuja kings could always bring their forces from one region to 
the other through this road and subdue them. Many stories of 
such unsuccessful rebellions are preserved in the local annals. 

The expansion of the Kambuja kingdom brought it into touch 
with the three important principalities in Burma which stood bet- 
ween it and the mainland of India. Ramannadesa, the country of 
the Raman or the Mon as they are called today, comprising the 
whole of Lower Burma, Tavoy, Mergui and Tennasserim, was the 
most powerful of these three and was something like a federation 
of states such as Ramavati, Hamsavati, Dvaravati, Sii Ksetra, etc. 
The number of these states varied, but was never less than seven, 
all acknowledging the suzerainty of one of them which grew more 
powerful than the others from time to time. It was a strong centre 
of Hindu civilisation, and contained a large number of famous 
colonies of Indians. To its north lay the kingdom of Pagan or 
Arimardanapura, in Upper Burma, along the valley of the Irawaddy 
and the Chindwin. Still further to the north and north-east, along 


the valleys of the Upper Irawaddy and the Salween rivers lay a 
number of Thai States which were often federated together and 
designated as Kausambl. The kingdom of Kambuja, occupying the 
central portion of Indo-Chinese peninsula, bordered all these three 
states which separated it from India. These regions have acquired 
a special importance in our days in view of the recent political 
situation. It is, therefore, interesting to note that they were never 
as remote and inaccessible from India as is generally supposed, and 
an unbroken series of Hindu or Hinduised kingdoms were spread 
over this vast area as far as the borders of China. Kambuja, the 
most powerful of all these states, not only established its political 
authority over the entire central part oi this vast region but intro- 
duced the elements of Indian culture and civilisation among the 
primitive peoples, mostly belonging to the Thai race. It is evident 
from a study of the Thai chronicles that even those Thai peoples 
or principalities which did not acknowledge the political suzerainty 
of Kambuja copied its civilisation and gradually imbibed the 
different elements of Hindu culture, traces of which, both literary 
and monumental, are still scattered over the entire area. This is 
a fascinating chapter of Indian colonial history but the limits of 
our subject do not permit us to go into further details. 

If we now turn from the north towards the south we find that 
Kambuja also came into contact with the mighty empire of the 
Sailendras in the Malay Peninsula. During the tenth century A.D. 
the northern part of this peninsula, lying, roughly speaking, to the 
north of the Isthmus of Kra, belonged to Kambuja, while the 
part to its south was included within the mighty empire of the 
Sailendras. We have no definite evidence of any political relation 
between the two, but Indravarman's claim of supremacy over Java 
may refer to a contest with the Sailendras who ruled over both 
Java and Malay Peninsula. 

Although we are unable to find out the exact relationship bet- 
ween Kambuja and the Sailendras, we are in a better position as 
regards her eastern neighbour, the kingdom of Champa. It will 
appear from what has been said above that almost throughout the 
ninth and tenth century A.D. there were perpetual hostilities 
between Kambuja and Champa, and Kambuja scored some definite 
successes against Champa in the tenth century A.D. 

This broad survey of the political condition of the Indo- 
Chinese Peninsula enables us to visualise how Kambuja grew to 
be a mighty power in Indo-China in the tenth century A.D. The 
history qLthe two succeeding centuries will show a further extent 


of its power and influence, but we must remember that the found- 
ation of this future greatness was laid by the dynasty of 
Indravarman and mainly by the exertion of himself and such 
powerful kings as Yasovarman and Rajendravarman. It is interest- 
ing to note in this connection that the Arab writer Ibn Al Fakih 
(902 A.D.) describes the Khmer kingdom as having an extent of 
4 months' march. 12 

The second characteristic that marks the history of Kambuja 
during the tenth century A.D. is the intensification of Hindu cul- 
ture. This may be clearly perceived from i, study of the numerous 
inscriptions, more than fifty in number, that this period has be- 
queathed. Most of these inscriptions are written in Sanskrit, 
in beautiful and almost flawless Kavya style, and some of them are 
quite big compositions. The text of an inscription of Yasovarman 
(No. 60) of which we possess no less than eleven copies in different 
places contains fifty verses. Another (Nos. 62-63) with five copies 
contains 108 verses each. A third (No. 61) contains 93 verses of 
which only 15 are common with the last scries. In addition to 
these there are a large number of records containing about iitty 
verses, a few more or less. But the prasastis of Rajendravarman 
exceed in size and quality even those of Yasovarman. The Mebon 
Ins. of this king (No. 89A) contains 218 verses, not a few of which 
are fairly big, being written in Saixlulavikrldita and Sragdhara 
metres. The largest inscription is that of Pre Rup (No. 93) which 
contains 298 verses. There are many other records of Rajendra- 
varman and other kings of this period which run to a considerable 

The authors of these inscriptions have very successfully used 
almost all the Sanskrit metres, and exhibit a thorough acquaintance 
with the most developed rules and conventions of Sanskrit rhetoric 
and prosody. Besides, they show an intimate knowledge of the 
Indian epics, Kavyas and Puranas, and other branches of literature, 
and a deep penetrating insight into Indian philosophical and 
spiritual conceptions; they are also saturated with the religious and 
mythological conceptions of the different sects of India; all this 
to an extent which may be justly regarded as marvellous in a com- 
munity separated from India by thousands of miles. It is beyond 
the scope of this lecture to illustrate these by citing examples from 
the different inscriptions. But a verse may be quoted to show how 

12. Ferrand~re;rte$ a p. 64. 


they were thoroughly conversant even with the grammatical treatise 
of Panini: 

Rajanvatity^zanyor-nrpo^invasat pmn 
Nlpdtandllaksanamantarena \ 

Padais^tu sadhutva-dhardn dharitrlm \ \ 

(No. 93, v. 48). 

This verse is a pun on the rule of Panini VIII 2*14 ( f rajanvan 
sawrdjye). It makes an exception to the general rule about the 
elision of the final n before the consonant in the word rajanvan in 
the sense of 'having a good king', the ordinary form 'rajavan' mean- 
ing only 'having a king'. There are similar references in an ins- 
cription of the time of Yasovarman: 

Sad-dharma-nirater~yyasya pada-rdjyena cakrire\ 
Upasarggdh kriyd-yoge te prdg dhdtor~mmuner~iva \ \ 

(No, 69, B. 13) . 

Here also the verse is a pun on Panini's Sutras 1,4,58,59,80 and com- 
pares king Jayavarman II with Panini. Similarly v. 15 of Ins. 62A 
compares king Yasovarman with Panini and all the epithets are 
applicable to both. The Mahabhasya was studied, and according 
to an Ins. of Yasovarman (No. 62D, D. 13) the king himself com- 
posed a commentary on it. The Minister of the king was an expert 
in Horasastra (No. 70, v. 8). Manu is mentioned as a legislator 
and a verse from Manu-smrti is reproduced verbatim (62A. C, 8 and 
9). Reference is also made to Vatsyayana, as the author of 
Kamasutra (62D. D. 1), and Visalaksa as having composed a 
treatise on Niti (62C, C. 15) . The famous medical treatise of 
Susruta is also mentioned in Ins. No. 61. (v. 49) . 

The Pre-Rup Ins. (No. 93) contains lour verses definitely allud- 
ing to Kalidasa's Raghuvamsa, repeating sometimes the very words 
used by the great poet. The inscriptions of Yasovarman refer to 
Pravarasena (No. 62B, B. 7) and Mayura (No. 62C, C. 16) as the 
authors, respectively, of Setubandha and Suryasataka, and to 
Gunadhya (Nos, 62C, C. 15; 62D, B. 26) as a writer in Prakrt with 
allusion to the legend about him contained in Kathasarit-sagara, 
The records frequently refer to the Trayl or Vedas, the Vedanta, 
Smrti, the sacred canons of the Budhists and Jainas, and religious 
texts of various Brahmanical sects and schools of philosophy. As 
to the Puranic religion and mythology and legends contained in 


Ramayana, Mahabharata and Harivamsa and the allusion, allitera- 
tion, simile, etc., usually met with in Sanskrit literature, one will 
come across them at every step as he proceeds through these 

These inscriptions bear ample testimony to the highly flourish- 
ing state of Sanskrit literature in Kambuja during this period. 
Unfortunately the Kambuja inscriptions are not generally known 
or studied in India. Otherwise they would have long ago been 
recognised as constituting an important addition to the Sanskrit 
Kavya literature of ancient India. It may be said without any 
hesitation that like the Kambuja monuments the Kambuja Sanskrit 
records on stone far exceed in volume and grandeur those of ancient 
India of which the existence is so far known. This is no mean 
compliment to the culture that the sons and daughters of India 
developed on the distant soil of Kambuja and spread far into the 
interior across the hills and dales of Laos and other inaccessible 
regions of the Indo-Chinese Peninsula. 

But apart from their literary merit these inscriptions are 
invaluable as testifying to the thoroughness with which Indian cul- 
ture and civilisation, in all its aspects, was imbibed in Kambuja, 
This is particularly applicable to the religious and spiritual life. The 
inscriptions give evidence of the minute knowledge of the rules, 
regulations and practices of religion, particularly of the aiva and 
Vaisnava sects, and show a thorough acquaintance not only with 
the various gods and goddesses in their numerous names and forms 
but also with the philosophical conceptions lying behind them. The 
prominent place occupied by religion in the life of the people is 
also demonstrated by the large number of temples and images 
erected and installed by kings and others. Most of the inscrip- 
tions refer to these pious foundations, and ruins of many of them 
are now lying scattered all over the country. But what strikes 
one more is that we find in Kambuja not only the external forms 
of Indian religion but that ethical and spiritual view of life which 
was the most distinguishing feature of ancient Indian civilisation. 
Anyone who carefully studies the inscriptions of Kambuja cannot 
fail to be struck with the spirit of piety and renunciation, a deep 
yearning for emancipation from the trammels of birth and evils 
of the world, and a longing for the attainment of the highest bliss 
by union with Brahma, which formed the keynote of their life 
and is expressed with beauty and elegance in language at once 
sombre and serene. 


Even the kings, high officials and the nobility of the kingdom 
were inspired by these high ideals. One of the interesting charac- 
teristics of the Kambuja court-life is the very intimate association 
between the secular and spiritual heads. The kings received their 
instruction in early life from eminent religious acaryas and there 
are many instances where sons of kings and members of the royal 
family became High Priests and Acaryas. The inter-marriage 
between the royal and priestly families was also a matter of fre- 
quent occurrence. The predominance of the priestly families who 
supplied royal priests for successive generations, such as we find 
in the Sdok Kak Thorn Inscription, already referred to above, is 
both an index and a cause of the spiritual outlook of the king and 
the people. The tutelary deity of the kingdom with the cult of 
Devaraja, placed in charge of a long line of High Priests who were 
the gurus or preceptors of the kings, must have helped to a great 
extent in moulding the whole view of life in the kingdom. 

But while all these causes undoubtedly operated in develop- 
ing the religious and spiritual life of the people, its main source 
must have been a close, constant and intimate contact with India. 
Fortunately this is not merely a hypothesis but may be proved by 
definite examples recorded in inscriptions of Kambuja. Rajalaksmi, 
the daughter of Rajendravarman, and the younger sister of Jaya- 
varman, was married to an Indian Brahmana Divakara Bhatta who 
was born on the bank of the river Kalindl or Yamuna sacred with 
the association of Krsna's boyhood (No. 103) . One of the ancestors 
of Yasovarman's mother, Agastya, is said to be a Brahmana of Arya- 
desa versed in Vedas and Vedangas (Nos. 60-62). Another 
Brahmana named Sarvajnamuni, versed in the four Vedas and all 
the agamas, and devoted to Siva, was born in Aryadesa (No. 170) . 
He came to Kambudesa and his descendants occupied high reli- 
gious office. There are many other less specific references to such 
migrations. 13 There is also evidence that the learned Brahmanas of 
Kambuja visited India. The most important instance is that of 
Sivasoma, the guru of Indravarman. We learn from an inscription 
(No. 58) that Sivasoma was the grandson of king Sri Jayendradhi- 
pativarman, maternal uncle of Jayavarman II, and that he learnt 
the Sdstras from Bhagavat-Sankara whose lotus-feet were touched 
by the heads of all the sages. It has been rightly conjectured by 

13. Ins. No, 161 probably refers to such a case. A learned muni was 
brought from a foreign land, which is not specified, in a fleet of barges, 


the editor of the Ins., that the reference here is undoubtedly to the 
famous Sankaracarya, 13a and presumably Sivasoma must have 
come to India to sit at the feet of the venerable Sankara. It may be 
noted in passing that as Indravarman lived towards the close of 
the ninth century A.D. givasoma must have flourished about the 
middle of the ninth century A.D., which agrees with the date 
generally assumed for Sankaracarya. 

The visit of Kambuja scholars to India may also be presumed 
on indirect evidence. M. Coedes, while editing the Vat Thipedi 
Ins. (No. 73), has pointed out that it exhibits all the characteristics 
of the Gauda style, described by Sanskrit rhetoricians, in such a 
striking manner, that its author must have either been born in 
Gauda or lived in that region. 

Though we can cite only a few actual instances of the learned 
Brahmanas of India, versed in sacred scriptures, settling in Kam- 
bujadesa, and the learned priests of the latter country visiting 
India, they corroborate what may be regarded as the only reason- 
able hypothesis that offers a satisfactory explanation of the 
Ihoroughness with which literary, religious and spiritual culture 
of India was imbibed by the people of Kambuja, 

It appears from the Kambuja inscriptions that the centres of 
Indian culture in Kambuja, from which it radiated all over tli3 
country, were the large number of dsramas which were founded 
by royal munificence and private efforts. Reference has already 
been made to these dsramas and the regulations concerning them 
in connection with king Jayavarman III. But they were very much 
developed during the peziod under review. According to Ins. 
No. 61 king Yasovarman founded one hundred excellent asramas 
in all parts of the kingdom. In spite of the possible exaggeration 
in number, there is no doubt that the king founded quite a large 
number of institutions of this kind. For the twelve diagraphic 
inscriptions (Nos. 60-61) undoubtedly refer to these asramas, 
called after the king Yasodharasrama, each being situated in a 
different locality and associated with a local temple containing a 

13a, Mr, S. Srikantha Sastri IMA, has opposed this identification, on 

grounds which appear to be very weak (IHQ. Vol. XVIII, pp. 175 ff). For 

arguments in favour of the identification cf. my paper in Indian Review 

(February, 194fl) and an article by Prof, K. A, Nilakanta Sastri (JOR. Vol XI, 

pp, 3-4). 


deity whose name is mentioned. Barring the passage containing 
the name of this divinity, the inscriptions are identical, and their 
most interesting part is that which contains elaborate regulations 
for these asramas, from which we quote the following extract. 

"All the things, which the king Yasovarman has given to the 
Asrama (Yasodharasrama) pearls, gold, silver, cows, horses, 
buffaloes, elephants, men, women, gardens etc., are not to be taken 
away by the king or anybody else. Into the interior of the royal 
cella the king, the Brahmans and the offspring of kings (ksatriyas) 
can alone enter without taking off their ornaments. Others, such 
as the common people forming the escort of nobles, can only enter 
in a humble dress without garlands the flower nandydvarta how- 
ever being allowed in their case too. (The common people) should 
not take any food or chew the betelnut there. The common people 
(not forming the escort of nobles) will not enter. There should be 
no quarrels. (Mock) ascetics of bad character should not lie down 
there. Brahmans, worshippers of Siva and Visnu, good people of 
good manners, can lie down there to recite their prayers in a low 
voice and to give themselves up to meditation. With the exception 
of the king whoever passes in front of the monastery shall get 
down from his chariot and walk uncovered by an umbrella. This 
is not applicable to strangers. The excellent ascetic, who is appoin- 
ted the head of the monastery, should always offer food, drink, 
betel and do all the duties, cs for example offering welcome to 
guests, such as Brahmans, children of kings (ksatriyas ) , ministers, 
the leaders of the army, ascetics of the Saiva and Vaisnava cult, 
and the best among the common people. They are to be honoured 
according to the order laid down here", 14 

As will be evident from these regulations, the dsramas to which 
they applied were open to all sects and classes of people. But three 
other inscriptions refer to dsramas specially meant for Vaisnavas, 
Saivas and Buddhists (Nos. 62A, 63, 63A) . These were called res- 
pectively Vaisnavasrama, Brahmanasrama, and Saugatasrama. The 
second name is somewhat puzzling, for in the regulations quoted 
above the term Brahmana appears as a general designation of a 
caste and there is nothing to indicate that it specifically refers to 
the Saivas alone. The regulations concerning these sectarian 
asramas are more detailed, and although there is much in common, 

14. The English translation is quoted from Chatter ji pp. 115-116 I have 
substituted -royal celia' for 'royal hut' in translating 'r 


there are some differences due to the emphasis laid on the special 
characteristics of the three religious sects. 

These dsrarnas were in charge oi a hulddliyaksa correspond- 
ing to the Kuiapati oi the previous ones and it is laid down at the 
beginning that lie and his assistants (karwakara) must observe the 
regulations (sdscma) laid down tor the particular asrama. llib 
lirst duty is to honour the guests and show all kinds oi hospitality 
to them, it the king comes there with the ladies oi his harem 
he shall entertain them as gods with all the resources at his com- 
mand. For, as laid down by Vyasa, tiie king ib the ouprerne lord 
oi the earth and the guru oi the entire world, arid whatever he 
wishes must be done. Next to the king, the Jbrahmanas shall be 
honoured above others, and if there is more than one, the pre- 
ference will be shown according to their character (sda), quahhca- 
tion (guna) and learning. Then he should honour in order of 
precedence, prince, minister, military chief and good men. Special 
nonour should be shown to the hero (sura) who has displayed 
bravery in battle. One who loves to fight shall be preferred to 
one who avoids it, for the defence of dhanna depends on the former. 

As to the persons next in order of precedence the regulations 
differ. In the case of Vaisnavasramas, iirst come those who know 
the three Vedas, and then the deary as versed 111 grammar, pre- 
ference among the persons of the same category being given to one 
who observes brahmacarya. Further, in preference to those who 
know the Pahcarutra or the grammar, honour should be shown to 
the teachers of these two sciences. 

In the Brahmanasrama, on the other hand, iirst the Brahmaiias 
and then the Saiva and Pasupata dcdryas should be honoured, pre- 
ference among them being given to the one who knows grammar. 
One who teaches Saiva and Pasupata doctrine or grammar is to be 
preferred to those who are versed in them. In the Saugatairama, 
too, the learned Brahmana should be honoured a little more than 
the dcdrya versed in Buddhist doctrine or in grammar, he who 
knows both being preferred to the other. The teacher of Buddhist 
doctrine or grammar should be preferred to one who merely knows 
these subjects. 

Next in order of precedence in all the three osramas is the 
learned grhastha, for learning is the best of the qualifications. 
Wealth, family, age, pious work and learning are the five titles of 
respect, each being superior to the one preceding it this is evi- 
dently a quotation from Manu II. 136. 


The Royal Decree then proceeds to say that the Kuladhyak$a 
shall give food, medicine and other necessaries to common people, 
particularly the boys, old men, poor, destitute and those suffering 
from illness, and worship the calf kapild by offer of grass etc. 

The Kuladhyaltsa shall also offer balls of rice and perform 
tarpana (funeral offering) at the parvan day, at the time of eclipse 
or at the end of each month, to the departed souls of those who 
were devoted or fell fighting or of those children, old men, poor 
and destitute, who had no relations to offer pmda. 

Those engaged in study in the asramas,~Vaisnavas, Brahmanas, 
or the Buddhist Bhikkhus, as the case may be, shall be supplied 
their daily necessaries, minute details of which are laid down with 
great care. 

If any innocent man seeks refuge in the asramas out of fear, 
he shall not be surrendered, and no one shall do any injury to 
another by words, thought or deed. Inoffensive animals shall not 
be killed within the boundaries of the dsrama. 

The daughter and grand-daughter of the king, old ladies of 
the royal family and the chaste women shall be honoured as guests, 
but they must not enter the monastic cells. Women of bad character 
shall not be permitted to enter even if they offer themselves as 
guests. In addition there is a rule in the case of the Buddhist and 
Vaisnava dsrama alone, that no inmate shall have any dealing with 
any woman, even though she may be his wife or co-religionist 
(sahardharmacdri) , within the environs of the dsrama. 

The regulations conclude with a list of officers and servants, 
specifying the service to which an adhydpaka or kulapati is entitled 
from them. Among the employees are two scribes, two keepers 
of books, and six men who prepared the leaves of manuscripts 
(patrakdraka) , showing that each dsrama was provided with a 
library (pustakdsrama) to which reference is made in other 

I have referred to these dsrama regulations at some length, for 
they reveal a picture of the social and religious life of Kambuja, 
such as we cannot get anywhere else. These dsramas were quite 
large in number, and spread over the whole country, and they served 
as strongholds or citadels of the Hindu culture and civilisation in 
its progress of conquest over the primitive culture of the land. 

There is one singular circumstance in connection with the ins- 
criptions setting forth these regulations, Some of them are digra- 


phic i.e., their texts are written in the normal Kambuja alphabet 
and again repeated in the North-Indian alphabet current at that 
time. Some of the texts are written in this latter script alone. 
No satisfactory explanation of this somewhat unusual feature has 
been forthcoming. It is reasonable to suppose that the text was 
written in North-Indian alphabet for the convenience of those who 
recently arrived from India and were not yet familiar with the 
Kambuja alphabet. They might have been pilgrims or ascetics 
who went there for short or long periods, if not for permanent 
settlement. For, regular or periodical visits of Indian religious 
teachers to Kambuja appear very likely. According to this hypo- 
thesis the digraphic inscriptions would be very strong evidence of 
an intimate relation between India and Kambuja. 



The death of Jayavarman V was followed by disputed succes- 
sion and civil war lasting for nearly ten years. The events that 
followed one another cannot be determined with absolute certainty, 
but the inscriptions discovered so far enable us to form a general 
idea of the situation. 1 

The immediate successor of Jayavarman V was Udayaditya- 
varman I who ascended the throne in 923 S (=1001 A.D.). The 
Ins. No. 117, dated in that year, gives a genealogy from which it 
appears that Udayadityavarman's mother, descended from the 
family of Sresthapura, was the elder sister of the queen of Jaya- 
varman V, and had an elder brother named Rajapativarman who 
was the senapati of the same king. This relationship is not such 
as would make us believe, without further evidence, that the 
succession was legitimate. That it was not peaceful is clear from 
the fact that we have smother king Suryavarman who issued 
inscriptions in the self-same year 923 (1001 A.D.) (No. 126) , 2 
To make the matter worse still, we find a thkd king Jayavlra- 
varman issuing inscriptions in the year 925 S (=1003 A.D.) 
(No 122). As a matter of fact Udayadityavarman I disappears 
from the scene altogether, under circumstances not known to us, 
from 1002 A.D. leaving the field to the two adversaries Suryavar- 
man I and Jayavlravarman. Suryavarman I ultimately came 
out victorious, but we have records of Jayavlravarman from 925 to 
928 g (~1003 to 1006) (No. 132). An analysis of the findspots of 
the inscriptions of these two kings shows that Jayavlravarman 
ruled in Angkor region and the regions to the west, while Surya- 
varman ruled in the north-eastern regions. An inscription found 
at Tuol Don Srei (No. 129) says of Suryavarman I that he carried 
on a war for 9 years and obtained the royalty in 924 S (=1002 
A.D,). The date 924, as that of his accession, is also given in many 
other records. 3 Now, it can hardly be accepted that the nine 

1. Cf. BEFEO. XXXIV, pp. 420 

2. Cf. also Ins. No. 117, v. 6, 

3. Nos. 148, 149, 


years 1 war preceded his accession, for in that case the war would 
have begun eight years before the death of Jayavarman V, whose 
reign appears from all accounts to be a peaceful one. The proba- 
bility, rather, is that the nine years' war refers to the struggle 
between Suryavarman and Jayaviravarman at the end of which, 
i.e. about 932 6, the former finally triumphed over his rival and 
enjoyed the undisputed monarchy of Kambuja. Why the year 924 
was chosen later as the formal date of his accession, although 
his reign began at least one year earlier in 923, it is difficult to 
say. Possibly Udayadityavannan died in 924 (1002 A.D.), and 
Suryavarman, who later tried to establish his claim to the throne 
by his relationship with previous kings, tried to pass himself as 
the legitimate successor of Udayaditya after his death. 

Nothing is known of the antecedents of Jayaviravarman. The 
inscriptions o* Suryavarman also refer to his origin in a somewhat 
vague manner. Iris. No. 148 says that he was born in the family of 
fndravarman. Ins. No. 146 repeats the same thing and add? that 
his queen Sri Viralaksmi was born of the royal line of Sii Harsa- 
varman and Sri Isanavarman. This is corroborated by No. 144. 
On the other hand Ins. No. 74B connects him with the maternal 
family of Indravarman. The Ins. No. 158 also probably refers to 
Siiryavarman as descended from the maTernal family of Jayavar- 
man V, but as some words of the verse are missing, one cannot be 
sure of the interpretation. 4 In any case these references show that 
Suryavarman had no legitimate claim to the throne by his relation- 
ship with any of his immediate predecessors, though he and his 
queen probably belonged to aristocratic families which claimed 
some relationship, however remote, with the royal family of 
Indravarman. The theory that Suryavarman originated from 
a ruling family in Siam or Malay Peninsula is an ingenious con- 
jecture but lacks convincing proof. 5 According to Ins. No. 149 

4. Cf. Corpus, p. 136, f.n. 4. 

5. The view was propounded by Coedes on the basis of a statement in 
C&madevlvamsa (a Pali chronicle) to the effect that the king of Kambuja, 
son of the king of Siridhammanagara, attacked Haripunjaya, about 20 years 
before the emigration of its inhabitants to Sudhammapura . As this event 
took place about 1056-7 A.D., the king of Kambuja must be Suryavarman. 
As Siridhammanagara is Ligor in the Malay Peninsula, it would follow that 
Suryavarman was a Malay prince. Coedes further points out that the title 
Kamtvan, introduced by Suryavarman, is derived from Malay tuan, a chief. 
This assumption also explains Suryavarman's leanings towards Buddhism, 


(y. 7) Suryavarman gained the kingdom by fighting with and defeat- 
ing its ruler, who was surrounded by other kings or m a state of 
intoxication. 6 This fully corroborates what we have said above. 
The same inscription refers to the king as being versed in Bha$yas, 
Kavya, six Darsanas and D;. s. 

This scholarly king seems to have been a Buddhist; for his 
inscription (No. 149) contains invocation to Buddha along with 
that to Siva, and his posthumous name was Nirvanapada. He 
issued edicts containing regulations about monasteries in which 
it was laid down that the ascetics and the Buddhist monk? should 
offer to the king the merits of their piety (No. 139) . But although he 
might have adopted the Buddhist faith, he did not give up the 
royal tutelary deity, and constructed both Saiva and Vai^nava 
temples. He is also said to have established the caste-system. 7 

There is no doubt that there was a prolonged civil war in Kam- 
buja during the early part of his reign which probably continued till 
1010 A.D. As a safeguard against similar outbreaks in future the 
king instituted a novel system which is known from ten inscriptions 
(No. 136), all dated 1011 A.D., eight of which are engraved on the 
pillars of the gopuram leading to the inner court of the royal 
palace of Angkor Thorn, and two on gateways of a neighbouring 
building. They contain the text of an oath, and the names of dis- 
trict officers, numbering more than four thousand, who took it in 
the presence of the sacred fire and the Brahmanas and the deary as 9 
offering unswerving and lifelong homage and allegiance to the king, 
and dedicating their lives to his service. 8 

Some of the expressions are very interesting. The officers 
swore that they "shall not honour any other king, shall never be 

indicated by his posthumous title Nirvanapada, for Ligor was a stronghold of 
Buddhism (BEFEO. XXV. pp. 24-5). This view, however, does not agree 
very well with the deduction made above from the findspots of inscriptions, 
viz., that Suryavarman was at first ruler in the north-east, while his rival, 
Jayavlravarman ruled over the western part of Kambuja, which is adjacent 
to Siam and Malay Peninsula. 

6. The meaning of the expression 'raja-samkirnna' is however very 
doubtful . 

7. No. 148, B. 8. The expression is 'varrmabMgc krte*. But as caste 
system is referred to in earlier records we cannot take the expression to 
mean that he introduced it for the first time. Possibly he re -organised it. 

8. Almost identical oath is taken by the royal officials of Cambodia even 
today on the occasion of the royal coronation (BEFEO. .XIII. 6. p. 16, and 
f.n. 3). 


hostile (to their king) and shall not be the accomplices of any 
enemy." These seem to refer to the recent civil war between 
Suryavarman and Jayaviravarman, though the inscriptions add that 
Suryavarmadeva has been in complete enjoyment of the sovereignty 
since 924 6 (1002 A.D.). Suryavarman seems to have con- 
quered the whole of Siam, and even carried his victorious campaign 
to the Mon Kingdom of Thaton in Lower Burma. But unfortu- 
nately no details are known. 8a 

Suryavarman was succeeded in A..D. 1049 by Udayaditya- 
varman II, who is said to have been born in a royal family. We 
dc not know definitely whether Udayaditya was the son of Surya- 
varman. According to the Lovek Ins. (No. 158) "when Surya- 
varman went to heaven, Udayadityavarman was crowned emperor 
(cakravartl) by his ministers." This rather implies that he owed 
the throne, not to rights of legitimate succession, but to the influ- 
ence of a party in the court. That perhaps explains the series of 
revolutions that harassed him throughout his reign. A graphic 
description of three rebellious outbreaks is given in right epic 
style in the Prah Nok Ins. (No. 153) which commemorates the 
valour and heroism of the general Sangrama who, being appointed 
the commander-in-chief for the protection of Rajalaksmi (royal 
fortune), stood by his king and defeated the rebels. 

The standard of revolt was raised in 1015 A.D. by Aravinda- 
hrada who made himself master of the southern half of the king- 
dom (v, 11). It is very likely that Aravinda was really a rival 
claimant to the throne, and contested it immediately after the death 
of Suryavarman; but being foiled in his efforts by the court-party 
he set up as an independent king in the south. In any case even 
the record of his enemy pays tribute to his military skill and the 
equipment of his army, and we are told that several distinguished 
generals sent against him proved unsuccessful. At last Sangrama 
asked for and obtained the permission of the king to lead the ex- 
pedition against the rebel Sangrama defeated the rebel who fled 
to Champa. 

The second rebellion was led by a favourite general of the 
king named Kamvau. Having collected a strong force, he overran 
the kingdom and defeated the royal army sent against him, after 

8a. This information is contained in Mon Annals (JBRS. XII, pp. 39 ff.; 
BEFEO, XXV, p. 24, 24). 


killing several renowned generals. At last Sangrama went in per- 
son to meet this dreaded foe. In the sanguinary battle that follow- 
ed, the two rival generals faced each other and Kamvau first 
struck Sangrma with an arrow. Sangrama, unperturbed, replied 
by shooting three arrows which pierced the head, the neck and 
the breast of Kamvau (v. 48) . He fell down and died, and his army 
was routed. 

The third rebellion was led by a chief called Slvat, who was 
aided by his yonger brother Siddhikara and another hero named 
Sagantibhuvana, each of whom prided himself as surpassing 
Kamvau in valour. But once more the general Saiigrama defeat- 
ed them and the rebel army took to flight. Sangrama pursued 
them as far as Prasanvrairmmyat but had to stop there to defeat 
some powerful enemy of the locality. After defeating him he 
again continued his pursuit of the rebel army of Slvat, and having 
bound the enemies in chains presented them to his royal master. 
He was richly rewarded by the grateful king for his unparalleled 
devotion an'd loyalty. 

This third rebellion was quelled in or about the year 1066 A.D. 
i.e. towards the very end of the king's reign. Thus practically the 
whole life of the king passed in great troubles. An echo of these 
troubles is reflected in several inscriptions. The Prasat Prah 
Khset Ins. (No. 154), for example, refers to the restoration of a 
linga that was damaged by the enemy Kamvau. This inscription 
also mentions a relation of king Udayarkavarman well-known in 
Madhyade6a. Udayarkavarman is no doubt the same as Udaya- 
dityavarman, but if Madhyadesa denotes the well-known region 
in India, it would follow that the king was connected with India. 

According to the Prasat Sralau Ins. (No. 156) the king desert- 
ed the town of Vrah Damnap. This is also evidently due to the 
troublesome rebellions. Udayaditya seems also to have sustained 
reverses in his wars against Champa. 9 Two inscriptions of 
Champa, dated 1050 A.D., refer to the glory of its king Jaya Para- 
mesvaravarmadeva as having penetrated into the Kambu king- 
<3om. Another inscription dated 1056 A.D. says that the king's 
nephew and general Yuvaraja Mahasenapati defeated the Khmers, 
took the town of Sambhupura and destroyed all its sanctuaries. 
The details of this campaign and the cause of the war are not known. 
The kingdom of Champa, at the beginning of the reign of Udaya- 

9. Champa pp. 146, 148-9, 155. 


iditya was passing through a series of troubles and considerably 
weakened by the recent Annamite invasions. It is not impossible, 
therefore, that the Kambuja king took the aggressive. But it 
appears that he sustained serious reverses as the Cham army ad- 
vanced up to the Mekong and pillaged the town of Sambhupura. 

The Sdok Kak Thorn Inscription (No. 151) , which records the 
glory of generations of royal priests from the time of Jayavar- 
man II, was engraved during the reign of Udyadityavarman II. It 
thus covers a period of two centuries and a half (802-1052) during 
which the same family supplied the high priests for the worship of 
the Royal God Devaraja. Jayendrapandita of this family was the 
guru of Udayadityavarman II, and taught him astronomy, mathe- 
matics, grammar, Dharmasastra, and all the other Sastras. 

The king had also another guru named Sankarapandita. The 
Lovek Ins. (No. 158) tells us that in imitation of the golden moun- 
tain of Jambudvipa where dwell the gods, he had a golden moun- 
tain built in the city, and consecrated a Siva-linga in a golden 
temple on the summit of the mountain. 10 Sankarapandita was 
appointed a priest of this linga, and apparently wielded a great 
influence in the state. For we are told that when Udayaditya- 
varman II died, Sankarapandita, along with the ministers, placed 
his younger brother (by the same mother) Harsavarman on the 
throne, and performed the ceremony of royal consecration. San- 
karapandita, of course, continued as royal priest. At the conclu- 
sion of the Lovek Ins. recorded by him, we are told that he was 
born, on the mother's side, in the family called Saptadevakula, and 
served three kings as priests. This implies that he was also the 
priest of Suryavarman. 

Harsavarman III ascended the throne in A.D. 1066. 11 The most 
important events of his reign are his expeditions to Annam. In 
1076 A.D. the Chinese emperor having decided upon an expedition 
against Annam, invited the rulers of Champa and Kambuja to 

10. This has been identified by P. Stern with the central tower of Bayon, 
but Coedes identifies it with Baphuon (BEFEO. XXXI. 18 ff.; XXVIII, 81 ff). 

11. Ins. No. 156 gives the date 987 S' (-1065 A,D.) for Harsavarman III, 
but according to Nos. 153 and 154 Udayadityavarman II was reigning in 
988 S', This apparent contradiction can be solved by supposing either (1) 
that the dates in the last two are current, and that in the first, an expired 
year, or (2) that Harsavarman III was proclaimed king during the lifetime 
of his brother, or (3) that one of the dates is wrong. The first alternative 
seems to be preferable (cf. Inscriptions, p. 223). 


help him. They sent military expeditions which retreated after 
the defeat of the Chinese. 12 Not long after this, hostility broke 
out between the kings of Kambuja and Champa. The details of 
1he campaign are briefly referred to in the Cham inscriptions as 
follows: "Harivarma (IV) (king of Champa) defeated the troops 
of Kambuja at Somesvara and captured the prince Sri Nandana- 
varmadeva who commanded the army." The battle must have 
taken place some time before 1080 A.D. 12a 

\ The foreign expeditions ending in defeat and disaster must 
have considerably weakened the power and prestige of the king 
und probably led to a political^ disintegration of the Kambuja 
kingdom. For although Harsavarman III continued to rule till at 
least 1089 A.D. (No. 159), we find another king Jayavarman (VI) 
issuing inscription in 1082 A.D. (No. 160). There is hardly any 
doubt, therefore, that towards the close of the eleventh century 
A.D., if not earlier still, during the troublesome reign of Udaya- 
dityavarman II, there were at least two rival kings in Kambuja. 
From an analysis of the fmdspots of inscriptions, so far discovered, 
it would appear that while Harsavarman III ruled in the Angkor 
region and to its south, Jayavarman VI ruled in the north and 
north-east. 13 This state of things probably continued till Surya- 
varman II, the second successor of Jayavarman VI, once more 
reunited the whole kingdom under his authority. 

Jayavarman VI does not appear to have been related in any 
way to his predecessors. The following genealogy of his family, 
whose ancestral home was Mahidharapura, 13a is furnished by the 
inscriptions Nos. 172 and 177. 

X ~ HiranvavarmanHiranyalaksmi 

D I I | ! 

| D Dharamndra- Jayavarman VI Yuvaraja 
I ! 

Ksitmdraditya-Narendralakmi Mahldharaditya-Rajapatmdralaksrni 
Suryavarman II ' * Harsavarman 

JayarajacudamaniDharanmdravarman II 

Jayavarman VII. 

12. BEFEO. XVIII, No. 3, p. 33. 
12a. Champa, p. 165. 

13. BEFEO. XXIX 299-300. 

13a. For a critical account of the dynasty of Mahldharapura cf BEFEO 
XXIX, p. 297. 


Hiranyavarman is called nrpa, mahipati and janesa, but whether 
he was an independent king or a vassal under Harsavarman or his 
predecessor is difficult to say. The fact that Jayavarman VI 
became king before his elder brother seems to show that in his 
case there was no question of a legitimate succession to the here- 
ditary kingdom, but acquisition of royal authority by a successful 
rebellion. It is very likely that Divakara Pandita who henceforth 
played a dominant part in the Court as the High Priest had, in 
some way, contributed to his success. He performed the coro- 
nation ceremony of Jayavarman VI and his two successors. 

Nothing is knoWn of Jayavarman VI. Ho died in 1107 A.D. 
and was succeeded by his older brother Dharanlndravarman I 
(No. 162). The latter being well advanced in age was unwilling 
to assume the burden of royalty, but had to yield to the wishes of 
the people, who were left without a protector at his brother's death. 
Dharanlndravarman was defeated by Suryavarman II, the 
daughter's son of his sister (No, 180), and the latter ascended the 
throne in 1113 A.D. (No. 165) . He is expressly said in two 
different Inss. (Nos. 165, 171) to have reunited the two kingdoms 
in Kambuja. It is obvious that one of these was ruled by Dhara- 
nmdravarman, and probably the other was under a descendant of 
Harsavarman III, both of whom were defeated by Suryavarman. 14 
Suryavarman was consecrated by Divakara Pandita who also 
initiated him into the mysteries of Vrah Guhya (the Groat Secret) , 
probably a tanlric cult. The king performed Kotihoma, Laksahoma, 
and the Mahahoma as well as various sacrifices to the ancestors 
(Nos. 167, 168). The cult of Bhadresvara, whose sacred temple 
was in Nfat Phu, the primitive capital of Kamhuja, seems to have 
come into porminence about this time (No. 170), Suryavarman 
further earned undying fame by constructing the famous Angkor 
Vat 14a one of the wonders of the world. Suryavarman sent two 
embassies to China in 1117 and 1121 A.D., thus resuming the dip- 
lomatic relations which were interrupted after the eighth century 
A.D. The Chinese Emperor conferred high titles on the Kambuja 
king whose dominions are said, in the Chinese annals, to have ex- 
tended from Champa to Lower Burma and included the northern 
part of Malay Peninsula up to the Bay of Bandon. This is corro- 

14. BEFEO. XXIX. p. 303. 

14a. This has been conclusively proved by G. Coedes (JA, lie serie 
Tome XV, 1920, pp. 96-100) , 


borated by other evidences. 15 Thus we see thai in spite of internal 
troubles the limits of the Kambuja empire exceeded even those of 
the tenth century. The Kambuja king is said to have maintained 
200,000 war elephants, and grandiloquent description is given of 
the royal tower, 16 

The Kambuja inscriptions refer in rapturous terms to tho 
victories of Suryavarman and his triumph over hostile kings, but 
they are very vague and do not give any specific information. He 
is said to have fought a terrible battle, in course of which he 
struck down his rival king from the head of his elephant and slew 
him (No. 171). This presumably refers to either Dharanmdra- 
varman or the king of the line of Harsavarman III who ruled over 
a part of Kambuja. We are further told that the kings of other 
islands whom he wanted to conquer voluntarily submitted to him 
and he himself marched into the countries of the enemies (No. 171) . 
These probably refer to his military campaigns in N. Annam and 
Champa, the details of which are preserved in Annamese Annals 
and the inscriptions of Champa. 

Suryavarman sent one or more expeditions to Champa and 
reduced the northern part of it, the kingdom of Vijaya, almost to 
a vassal state of Kambuja. In 1128 he sent an expedition, 20,000 
strong, against the kingdom of Annam, and invaded Nghe-an. 
At the same time he sent a fleet of 700 vessels, which with the sup- 
port of the army of Champa was to attack Hatin. The army cross- 
ed the Annamite Hills by the pass of Ha-trai and descended to the 
valley of Pho-giang. There they halted and waited for the fleet 
and the Cham army. But these did not arrive in time, and in the 
meantime the Annamite troops fell upon the Kambuja army and 
defeated it. The Kambuja general was killed and his army 
beat retreat. The fleet and the Cham army arrived after a few 
months, and deprived of the support of the Kambuja army, re- 
treated, after having ravaged the coast of Nghe~an and 
Tnan-hoa. In 1132 A,D. the army of Kambuja, with that of 
Champa, again invaded Nghe-an, but was easily repulsed by the 

15. For the Chinese account cf. Ma -T wan-Lin's Meridional* x translated 
by Hervey de St. Denis, p. 485. The bas-reliefs of Angkor depict Siamese 
soldiers as fighting under Kambuja generals (BEFEO, XXV, p. 18; JHQ, 
I, p. 618; BCAL 1911, p. 103), 

16. These and other details are given in the Chinese account (see pre- 
ceding footnote). 


governor of Than-hoa. The two countries then made peace with 
Annam and sent an embassy in 1135. Two years later, the Kam- 
buja army again invaded Nghe-an and was again easily repulsed. 
The Cham army did not join the Kambujas in this last expedition 
and observed the peace with Annam. 17 Probably to punish this 
desertion or taking advantage of the accession of a new king, Jaya 
Harivarman, in the southern part of Champa, Suryavarman decided 
to complete the subjugation of Champa. A detailed account of the 
campaign is given in inscriptions of Champa, which may be summed 
up as follows. 18 

Scarcely had the king ascended the throne when the king of 
Kambuja commanded Sankara, the foremost among his generals, 
to go and fight him in the plain of Rajapura. Sankara was aided 
by a large number of troops from Vijaya i.e. the portion of Champa 
subject to Kambuja. Harivarman met the hostile army at Chak- 
lyan (probably the village of Chakling in the southern part of the 
valigy of Phanrang, in the neighbourhood of the rock of Batau 
Tablah which contains an inscription describing the battle) and 
gained a great victory. As the Myson inscription tells us: "Jaya 
Harivarman fought against the general Sankara and all the other 
Cambodian generals with their troops. They died in the field 
of battle." This happened in 1147 A.D. Next year "the king of 
Kambuja sent an army thousand times stronger than the previous 
one to fight in the plain of Virapura". Harivarman met them at 
the field of Kayev and completely defeated them. 

Having defeated the two armies sent against him, Harivarman 
now felt powerful enough to take the offensive. The king of Kam- 
buja did not underrate the danger. He hastily consecrated Hari- 
deva, the younger brother of his first queen, as king of Vijaya, and 
"commanded various generals to lead the Cambodian troops and 
protect prince Harideva untilj he became k^ng in the city of 
Vijaya." Jaya Harivarman also marched towards that city and 
probably re-took it before the arrival of Harideva. In any case 
the two hostile armies met at the plain of Mahisa "to the east of the 
temple of Guhesvara on the river Yami," and Harideva was defeat- 
ed and killed. "Jaya Harivarman destroyed the king Harideva 
with all his Cham and Cambodian generals and the Cham and 
Cambodian troops; they all perished." 

17. BEFEO, XVHE, No. 3, p. 33. 

18, Champa, Part I pp. 96 ff; Part II, pp. 179-80, 193, 


This battle took place in 1149 A.D. 18a Next year the Kambuja 
king sent an expedition against Annam. The Kambuja army 
started in autumn, and the monsoon proved disastrous. The 
troops suffered heavily from fever during their passage over the 
Annamite hills and they were so weakened when they reached 
Nghe-an, that they retreated without any fight. 19 

Thus the foreign expeditions of Suryavarman proved disas- 
trous and it had serious repercussion on the fortunes of Kambuja. 
The history of this kingdom after the death of Suryavarman II is 
obscure, but it is definitely known to have suffered much both 
from internal dissensions and the series of defeats inflicted by the 
king of Champa, 

The last known date of Suryavarman II is 1145 A,D. though 
he probably ruled for some years more. He had the posthumous 
title Paramavinuloka. 19a He was probably succeeded by Dharan- 
indravarman II, the son of his maternal uncle, though we 
are not quite certain about it. Dharanmdravarman's queen was 
the daughter of king Harsavarman, who is most probably to be 
identified with king Harsavarman III. 20 We do not know anything 
about Dharamndravarman II, not even whether he was the legiti- 
mate heir to the throne or got it by peaceful means. The same 
uncertainty prevails about the next king Yasovarman II, though 
we know some details of his reign from an inscription of Jaya- 
varman VI^ found at Bantay Chmar. 21 It tells us that Yasovar- 
man was faced with a rebellion headed by Bharata Rahu Sambuddhi. 
Evidently the rebellion assumed at one time serious proportions, 

18a. As the last known date of Suryavarman II is 1145 A.tX a part of this 
campaign which ended m 1149 A.D., and the subsequent troubles created in 
Champa by the Kambuja king (Champa, pp. 97-8) might belong to the next 
reign or reigns. 

19. BEFEO, XVIII, No. 3. p. 34. 

I9a. Cf JA. lie Serie, Tome XV, 1920, pp. 96-100 

20. BEFEO, XXIX, p. 327. 

21. King Yasovarman mentioned in the Bantay Chmar Ins. (No. 182) 
was identified by Aymonier (II. pp. 344-5) with Yasovarman I and the revo- 
lution and campaign against Champa, mentioned in this record, were placed 
during his reign. This view was followed by all the writers on the subject 
until Coedes demonstrated that the events related in the record refer to a 
new king Yasovarman II of the 12th century A,D. For the account that 
follows of the two hitherto unknown kings, Yasovarman II and Tribhuva- 

nadityavarmnan, and prince Srmdra-Kumara cf. Coedes' brilliant exposition 
in BEFEO XXIX, pp. 305 ff. 


for we are told that the rebels attacked even the palace itself and 
the royalist troops in the capital took to flight. Prince Srlndra- 
kumara, 2:i son of the future king Jayavarman VII, came to the 
rescue of the king. He himself fought in person with the rebels 
and defeated them. 

It is in connection with this fight that we come across the term 
Sanjak, which presumably means a chief bound by a special 
oath or obligation to defend the person of the king or a prince, 
When Srindrakumara came out to fight with the rebels, his body 
was covered by two Sanjaks, who were killed before his very eyes. 
The king showed appreciation of their service in a befitting man- 
ner. Posthumous honours were bestowed on them and their 
statues were installed in shrines. The fact that the inscriptions 
refer to these statues as gods shows that they were deified like 
kings. Needless to add that the king bestowed wealth, favours 
and honours on the members of their families. 

The rebellion did not seriously affect the solidarity of the 
kingdom. For Yasovarman felt powerful enough to send an ex- 
pedition against the kingdom of Champa, led by the same prince 
Srindrakumara. At first his enterprise proved successful. He 
seized the fort which Jaya Indravarman, king of Champa, had 
built on Mount Vek, and placed a Cham general on the throne of 
that kingdom. But the re-organised Cham troops caught Srindra- 
kumara in an ambush and surrounded him. He was forced to re- 
treat, though continually fighting with his enemy, and took shelter 
on Mount Trayachur, The Cham troops besieged him on the 
mountain, but the prince fought his way down to the foot of the 
hill, and succeeded in breaking through the enemy line. On this 
occasion, too, he owed his life to the deliberate self-sacrifice of two 
of his Sanjaks who gave their lives in defending his person. As 
on the previous occasion, the king conferred posthumous honours 
on the two heroes and installed their statues in a shrine. Though 
the prince safely returned with his army to his kingdom the whole 
expedition was an ignominious failure. 

The prince Srmdrakumara died while young, and his statue 
also was placed in the same shrine where those of his four faithful 
Sanjaks were installed. 

22. Sri is a part of the name, and not an honorific 


The war with Champa, however, continued, and another ex- 
pedition was sent to Vijaya (central Champa) , probably under the 
future king Jayavarman VII. About this time, and evidently 
taking advantage of the absence of royal troops in Champa, another 
rebellion took place in Krunbuja, headed by Tribhuvanadityavar- 
man. As soon as he heard of this outbreak, Jayavarman returned 
with his troops to Kambuja, but he was too late. For Yaiovar- 
man II was defeated and killed and Tribhuvanadityavarman had 
ascended the throne of Kambuja. This took place before 1166 
A.D. Jayavarman evidently could do nothing and simply bided 
his opportunity which was not long in coming. 

The facts recorded ^bovc prove that Yasovarrnan II was con- 
n^cted with the royal family, for otherwise it is difficult to account 
for the loyal services rendered to him both by Jayavarman and his 
son Srlndrakumara. Tribhuvanadityavarman, on 1he other hand, 
appears to be a usurper and adventurer, having no connection with 
the royal family, but merely an official (blirtya) . This new king 
of Kambuja was involved in a prolonged fight with Champa, with 
disastrous consequences. Fortunately, this campaign is referred 
to not only in the inscriptions of Kambuja (Nos. 180, 181) but also 
m the Chinese annals, 23 and both represent Jaya Indravarman, 
king of Champa, as the victor. According to the Kambuja ins- 
cription Jaya Indravarman invaded Kambuja with a big army and 
Tribhuvanadityavarman was defeated and killed. But the Chinese 
account gives some details of the campaign. We are told that 
Jaya Indravarman invaded Kambuja in 1170 A.D. (or three years 
earlier) 24 and the war went on for seven years without any decisive 
result. At last the Cham king equipped a fleet and sent a naval 
expedition in 1177 A.D. The fleet sailed up the Mekong river and 
reached the capital city, and Jaya Indravarman plundered the 
capital and then retired, carrying an immense booty with him. 
According to the Kambuja inscription the Kambuja king, evidently 
Tribhuvanadityavarman, was killed in this fight, 25 but Kambuja 

23. cf. BEFEO. II, p. 130 and Maspero Le Royaume de Champa p. 164, 
where full references are given. The account is also given in Champa, pp. 

24. Maspero (op. cit.) gives the date on Chinese authority but the Po 
Nagar Ins. (in Champa) of Jaya Indravarman, dated 1167 A.D., refers to 

his expedition against Kambuja (Champa, Part II. p. 198). 

25. This is also mentioned in one Chinese account (BEFEO. II, p. 130) 
though Maspero discredited it on grounds, which do not appear to be quite 
strong (op. cit. p. 164, f.n. 6). 


was saved by the heroism of Jayavarman. He defeated the Chams 
in a naval engagement 26 and made himself master of the kingdom 
of Kambuja, four years later. 

With the accession of Jayavarman VII in A.D. 1181 (No. 178) 
we are again on the firm ground. He was the last great king of Kam- 
buja and we know a great deal about him, his military campaigns, 
his religious foundations and his woi^ks of public utility. 

As regards the first, he attained conspicuous success in his 
wars with Champa, the eternal enemy of Kambuja. As already 
noted above, he had probably fought in Champa on behalf of Yaso- 
varman II. The war was renewed after he had ascended the 
throne. According to the Ins. No. 177 (v. 28) he took the Cham 
king prisoner and then released him. According to the Chinese 
account of Ma Twan Lin, Jayavarman fully avenged the sack of 
the Kambuja capital by the king of Champa. The King of Kam- 
buja, we are told, invaded Champa, dethroned its prince, and put 
one of his own men in his place, and for long Champa remained a 
vasspl state of Kambuja. 27 

Fortunately the Inss. of Champa supply us very interesting and 
detailed information about the war between Champa and Kam- 
buja 28 which was begun in 1190 A.D. by the aggressive campaign 
of Jaya Indravarman on Vatuv king of Champa. The Kambuja 
king Jayavarman VII sent an expedition to check his advance and 
invade Champa. 

The leader of this expedition, who was ultimately destined 
to play an important part in history, was Sri Suryavarmadeva, 
prince Sri Vidyanaridana. He was apparently an inhabitant of 
Champa, but betook himself early in life to Kambuja (1182 A.D.) . 
The king of Kambuja welcomed him and employed his services on 
occasions. Thus we read: "During his stay at Kambuja, a 'depen- 
dent town of Kambuja called Malyan, inhabited by a multitude of 
bad men, revolted ageinst the king of Kambuja. The latter, seeing 

26. The scenes of this naval victory are perhaps depicted in the sculptures 
of Bayon and Bantay Chmar (BEFEO. XXIX, p. 326). 

27. Cf. also BEFEO. It. 130 for other Chinese references to this. The 
date of his expedition, as given in the Chinese accounts, is not correct (Mas- 
pero-op. cit. p. 164. f.n. 8). 

28 Cf. Champa, pp. 106 ff. from which the following account is quoted. 
The passages within quotation marks are extracts from Cham Inscriptions, 


the prince well-versed in arms, ordered him to lead Ihe Kambuja 
troops and take the town of Malyan. He did all that the king of 
Kambuja desired." 

The king of Kambuja, pleased at his valour, conferred on him 
the dignity of Yuvaraja, and when war broke out with Champa, 
as related above, he "sent the prince at the head of Kambuja 
troops in order to take Vijaya, and defeat the king Jay a Indravur- 
man Oh Vatuv. Sri Suryavarmadeva obtained a complete victory. 
He captured the king of Champa and had him conducted to Kam- 
buja by the Kambuja troops." 

The king of Kambuja now divided Champa into two portions. 
He placed his own brother-in-law, Surya Jayavarmadeva prince 
In, as king of the northern part with Vijaya as capital, while 
Suryavarmadeva, prince Sii Vidyanandana, the victorious general, 
became king of the southern portion with his capital at Rajapura 
in Panran. 

Suryavarmadeva prince Sri Vidyanandana defeated a number 
of 'thieves or pirates', apparenlly the adherents of the late regime 
that had revolted against him, and reigned in peace at Rajapura. 
The northern kingdom, however, was soon lost to Kambuja. 
Within two years, prince Rasupati, apparently a local chief, led a 
revolt against the Kambuja usurper, Sri Surya Jayavarmadeva 
prince In. The latter was defeated, and returned to Kambuja, 
while Rasupati ascended the throne under the name of Sri Jaya 

The king of Kambuja now sent an expedition against Vijaya 
(1192 A.D.). With a view, probably, to conciliate the national 
sentiments by placing the captured king of Champa Sri Jaya 
Indravarman Oh Vatuv on the throne, as a dependent of Kambuja, 
he sent him along with this expedition. The Kambuja troops first 
went to Rajapura. There the king Suryavarmadeva prince Sri 
Vidyanandana put himself at their head, and marched against 
Vijaya. He captured Vijaya and defeated and killed Jaya Indra- 
varman (Rasupati). 

The victorious king of Rajapura now ascended the throne of 
Vijaya and the whole of Champa was again re-united under him. 
Jaya Indravarman Oh Vatuv, the king of Champa, being deprived 
of the throne, fled to Amaravatl. There he collected a large 
number of troops and advanced against Vijaya. "The king defeat- 
ed him, compelled him to fall back on Traik, and there captured 


him and put him to death/' Henceforth Suryavarmadeva Vidya- 
nandana ruled over the whole of Champa without opposition 
(1192 AD.). 

But he had shortly to reckon with the king of Kambuja whom 
he had so basely betrayed. In 1193 an expedition was sent 
against him, but he gained an easy victory. Next year the expedi- 
tion was repeated on a larger scale. 

"In Saka 1116 (1194 A.D.) the king of Kambuja sent a large 
number of generals with all sorts of arms. They came to fight with 
the prince. The latter fought at Jai Ramya-Vijaya, and vanquish- 
ed the generals of the Kambuja army." 

But the king was not destined to enjoy his sovereignty for a 
long time. He was defeated in 1203 A.D. by his paternal uncle, 
called Yuvaraja (or son of Yuvaraja) Mnagahna Oh Dhanapati or 
Yuvaraja On Dhanapati grama, who was sent by the king of Kam- 
buja against him. 

The career of this Yuvaraja was analogous in many respects 
to that of king Suryavarman himself. He, too, lived as an exile 
in the Court of Kambuja and obtained the favours of the king by 
successfully suppressing the revolt of Malyan. It is just possible 
that these two Cham chiefs, uncle and nephew, both went together 
to Kambuja and the Malyan revolt, which both claim to have sub- 
dued, was the self-same military expedition in which both of them 
took part. But the nephew soon surpassed the uncle and, as we 
have seen above, ultimately became the king of Champa. 

The king of Kambuja, twice baffled in his attempt to defeat 
him, at last sent the uncle against the nephew. In 1203 A,D. king 
Suryavarman was defeated and the Yuvaraja On Dhanapati ruled 
over Champa. 

The Yuvaraja Dhanapatigrama, who now ruled over Champa, 
had a hard time before him. Rebellion broke out in various parts 
of the kingdom. The most formidable was one led by Putau 
Ajna Ku, but it was put down by the Yuvaraja. "Then Putau 
Ajiia Ku revolted. He conquered from Amaravati as far as 
Pidhyan. The king of Kambuja commanded the Yuvaraja to lead 
the troops of Kambuja and capture Putau Ajna Ku. He captured 
him and sent him to Kambuja according to the desire of the king". 

The king of Kambuja, pleased at his valour, conferred high 
dignities on him, and apparently formally appointed him as the 
ruler of Champa in 1207 A.D. 


It appears, however, that Champa was at this time very hard 
pressed by the Annamites. The Cho Dinh Inscription tells us: 
"Then (sometime after 1207 A.D.) the Siamese and the Pukam 
(people of Pugan in Lower Burma) came from Kambuja and a 
battle took place with the Annamites. The Kambuja generals led the 
troops which opposed the Annamites and the loss on both sides was 
very great." The Annamite documents inform us that the Chams 
aided by the Cambodians attacked Nghe-An in 1216 and 1218, but 
the governor of the Province dispersed them. It would thus appear 
that since about 1207 A.D. a series of battles followed, in which 
victory more often inclined to the Annamites. 

These long-drawn battles must have exhausted the Kambujas, 
As a matter of fact, the series of warfares in which they were in- 
volved ever since 1190 A.D. when they conquered Champa, must 
have proved too great a burden for the people and, to make matters 
worse, the Thais in Siam at this time began to press them hard 
from tho west. At last in 1220 A.D. the Kambujas evacuated 
Champa, and a formal peace was probably concluded with Ansa- 
raja of Turai Vijaya in 1222 A.D. 

It is not definitely known whether Jayavarman VII was still 
alive when this final withdrawal took place. Most probably he had 
ceased to rule some time before it, for we do not know, even 
approximately, the year of his death, 29 

But the credit of conquering Champa and making it a vassal 
state of Kambuja belongs to him. It was the crowning triumph at 
the end of an age-long struggle which extended the frontier of the 
Kambuja empire to the China Sea on the east. But Jayavarman 
VII did not confine his military activities to the eastern frontier of 
his kingdom. On the western side, too, he appears to have attain- 
ed some success against the kingdom of Pagan. Since the middle 
of the eleventh century A.D. this kingdom grew very powerful and 
subjugated Ramannadesa, thereby extending its authority over the 
whole of central and southern Burma. According to the Chinese 
chronicles 30 Pagan was annexed to Kambuja towards the close of 
the twelfth century A.D, Whether the Chinese mean the whole 
of Pagan or some part of it adjacent to Kambuja empire, such as 

29. The view, generally accepted, that Jayavarman VII died in 1201 A.D., 
has been proved to be wrong by Coedes (BEFEO. XXV. pp. 393-4; XXIX t 
p. 328). 

30, Aymonier, III, 528. 


Pegu, we cannot say. But in any case it shows a further expan- 
sion of the Kambuja empire beyond its western limits in the tenth 
and eleventh century A.D. The credit of this conquest certainly 
belongs to Jayavarman VII, though unfortunately we have no 
details in his inscriptions beyond a vague reference that he defeat- 
ed a king (No. 179). It must be remembered, however, that the 
Cham inscriptions, noted above, refer to the people of Pukan 
(Pagan) as having served in the Kambuja army. Ins. No. 177 also 
says that the men of Champa and Pukan were employed as servants 
in the temples. No doubt they were mostly prisoners from Burma 
and Champa. The Ceylonese chronicle Mahavamsa has preserv- 
ed a story to the effect that a princess sent by king Parakrama- 
bahu of Ceylon (1164-1197 A.D.) to Kambuja was seized by the 
Burmese king who also imprisoned the Ceylonese envoys on the 
ground that thoy were sent to Kambuja. 30a This seems to be an 
echo of the hostility between the kings of Kambuja and Pagan. 

Jayavarman VII thus ruled over an empire which stretched 
from the Bay of Bengal to the Sea of China. The central regions 
in the Indo-Chinese Peninsula and the northern part of Malay 
Peninsula formed part of his empire as in the tenth century A.D. 

The religious foundations and works of public utility under- 
taken by Jayavarman VII were on a scale befitting the mighty 
empire over which he ruled. The account of royal donations con- 
tained in the Ta Prohm Ins. (No. 177) makes interesting reading 
and reveals the magnitude of the resources and depth of religious 
sentiments of the king. It concerns the Rajavihara i.e., the temple 
of Ta Prohm and its adjuncts where the king set up an image of 
his mother as Prajfia-paramita. It is not possible here to record 
all the details but a few facts may bo noted. Altogether 66,625 
persons were employed in the service of the deities of the temple 
and 3,400 villages were given for defraying its expenses. There 
were 439 Professors and 970 scholars studying under them, making 
a total of 1409, whose food and other daily necessaries of life were 
supplied. There were altogether 566 groups of stone house? and 
288 groups of brick. Needless to say that the other articles, of 
which a minute list is given, were in the same proportion, and 
they included huge quantities of gold and silver, 35 diamonds, 
40,620 pearls and 4,540 other precious stones. All these relate to a 

30a. MaMvamsa, ch. 76, vv. 21-35. JASB. XII, pp. 197-201; BEFEO. H, 
p. 130. 


single group of temples. And the inscription informs us that there 
were 798 temples and 102 hospitals in the whole kingdom, and 
these are given every year 117,200 khdril^ds of rice, each khdriha 
being equivalent to 3 maunds 8 seers. In conclusion the king ex- 
presses the hope that by his pious donations, his mother might be 
delivered from the ocean of births (bhavdvdhi) . 

Of the 102 hospitals mentioned above, the site of 15 can be 
determined by means of inscriptions which record their founda- 
tions. 31 These inscriptions are almost identical and lay down 
detailed regulations about the hospitals. They give us a very good 
idea of the system of medical treatment organised by the 
state, 32 but require separate treatment and cannot be discussed 

Jayavarman VII also established 121 Vnlmigrlw or dharam- 
solas along the principal routes within his kingdom for the con- 
venience of pilgrims and other travellers. 33 

Jayavarman VII was a devout Buddhist and received the 
posthumous name Maha-parama-saugata. Many of his records ex- 
press in beautiful language the typically Buddhist view of life, 
particularly the feelings of charity and compassion towards the 
whole universe. 

Jayavarman VII reigned for more than twenty years, His 
death which nearly coincides with the closing of the twelfth cen- 
tury A.D. constitutes an important landmark in the history of 
Kambuja. For it is now generally recognised that the eleventh 
and twelfth centuries A.D. form the most glorious period in the 
annals of the great Hindu colony. It saw the greatest expansion 
of the Kambuja empire which extended from Lower Burma to 
Annam, and reached the Bay of Bengal on the west and the China 
Sea on the East. In the north, most of the Thai principalities in 
Laos acknowledged its suzerainty and its boundaries touched those 
of the Chinese Empire, In the south it not only comprised the 
whole of Siam, Cambodia and Cochin-China, but also a part of 
Malay Peninsula. The wild peoples of the comparatively inaccessi- 
ble uplands of Laos were brought under the civilising influence of 

31. BEFEO, XL, p. 344. 

32. BEFEO, m, p. 18. 

33. BEFEO. XL. p. 347, 


the Hinduised Khmers, and the onward march of Hindu culture 
followed in the wake of the victorious campaigns of the Kambuja 
kings. The glory of the Hindu culture is seen at its best in these 
palmy days of the Kambuja Empire. The cultivation of Sanskrit 
literature, both secular and religious, reached the highest develop- 
ment in Kambuja during this period, and the themes of the two 
great Indian epics adorned the walls of its most famous monu- 
ments, High priests, learned in Hindu and Buddhist canons, ob- 
tained a dominant position in state and society, and we possess 
elaborate and lengthy records of quite a large number of eminent 
Brahman families. They typified the essence of Indian culture and 
were mainly instrumental in transplanting and diffusing the higher 
intellectual, moral and spiritual life of India on the soil of Kam- 
buja. The long and elaborate pmsastis of the Kambuja kings, 
though poor in respect of data for political history, are rich in 
supplying materials for the reconstruction of the spiritual life of 
the people. The kings and grandees of Kambuja, not to speak of 
its priests and dcaryas housed in the hundreds of asramas reared 
by the piety of the rulers and nobles, were imbued with the 
impulses of that higher life in man which forms such a distinct 
characteristic of Indian civilisation. It is not merely the externals 
of Hindu civilisation but the very essence of the Hindu view of 
life that is unfolded before our eyes as we study the records of 
this period, and review the achievements of the greatest kings of 
Kambuja. What, for example, can be a nobler sentiment than that 
which inspired king Jayavarman VII in founding the hospitals, 
more than 100 in number, all over his kingdom? In the record of 
the foundations we read: 

"Dehinam deha-rogo yan-manorogo rujattaram \ 
Rastra-duhkham hi bhqrtrnam duhkhan^duhkhantu 
natmanah \ \ 

"The bodily pain of the diseased became in him (king Jaya- 
varman VII) a mental agony more tormenting than the former. 
For the real pain of a king is the pain of his subjects, not that of 
his own (body)." 

This noble sentiment, which combines the idealism of the 
Kautilyan king with the piety and humanity of Asoka, was not a 
mere pious wish or thought but was actually translated into action 
by the elaborate system of remedial measures with a network of 
102 hospitals as its nucleus, 


The eleventh and twelfth centuries A.D. also produced the 
most notable monuments in Kambuja which still excite the wonder 
and admiration of the world. It has not been possible for me to 
review the history of art in Kambuja in course of these lectures, 
because its evolution and chronology is still a matter of dispute 
and the subject therefore requires a separate and detailed treat- 
ment in a series of lectures wholly devoted to it. But viewed as 
a whole, the monuments of Kambuja, both by their massive charac- 
ter and unparalleled grandeur, form the most brilliant testimony 
to the richness and splendour of a civilisation of which the written 
records give but an imperfect picture. It would be no exaggera- 
tion to say that in respect of architecture the Indian colony of 
Kambuja far surpasses the mother-land, even though the achieve- 
ments of the latter are by no means of mean order. In spite of 
uncertainties of chronology we may refer a few monuments, with 
tolerable certainty, to this period which constitutes in so many 
ways the most memorable one in the history of Kambuja. 

The most famous of the monuments of Kambuja, viz., Angkor 
Va4 was built by king Suryavarman II who ruled between 1113 
and 1145 A.D. The Baphuon, another noble monument, was for- 
merly referred to the 9th and 10th centuries A.D. but is now re- 
garded as belonging to the reign of Udayadityavarman II (1049- 
1066 A.D.), 34 The famous Angkor Thorn, with its gate-towers, 
ramparts and ditches, and the Temple of Bayon in the centre of 
the city, were formerly attributed to Yasovarman I (889-908 A.D.) 
but are now believed by some to be the work of Jayavarman VII, 35 
who ascended the throne in the year 1181 A,D. Another famous 
monument, that of Bantay Chmar, whch was formerly attributed to 
Jayavarman II (9th century) is also referred by some to Jayavar- 
man VII 36 and others to Yasovarman II 37 (1160-1180 AD.). In 
short, whereas the majority of the splendid monuments of Kam- 
buja were formerly placed in the 9th and 10th centuries A.D., 
their date is now brought down by nearly two or three hundred 
years, and instead of Jayavarman II and Yasovarman I, the four 
successive kings Suryavarman II, Dharamndravarman II, Yaso- 
varman II and Jayavarman VII, whose reigns practically cover the 

34. BEFEO, XXXI, p. 18. 

35. BEFEO. XXVm, p. 81. 

36. Ibid, p. 100. 

37. BEFEO. XXXV 5 p. 181, 


whole of the twelfth century A.D., appear to be the great builders 
of Kambuja monuments. We must, therefore, give up the old idea 
that the twelfth century was a period of decay in the history of 
Kambuja, and rather regard it as the period of the greatest glory 
of Kambuja, 

It is not possible in the present course of lectures to give such 
a detailed description, even of the most famous monuments of 
Kambuja, as would convey a fair idea of their nature and artistic 
excellence. I would therefore merely attempt to indicate, in a 
general way, the special features which characterise them. 

The earliest series of monuments at Angkor consist of isolated 
temples which show great resemblance with Indian temples. But 
gradually a new style is evolved in the eleventh and twelfth cen- 
turies A.D., first by the introduction of gallery, and later still, by 
pyramidal construction in several stages. The combination of 
these two features results in a series of concentric galleries, en- 
closing each successive stage of the pyramid, with a crowning 
tower at the centre of the top or the highest stage. Similar towers 
are added at the four corners of each stage of the pyramid, and 
finally we have the gopurams at one or all the four faces, each 
consisting of a gateway with a vestibule, surmounted by an orna- 
mental tower in the form of a stepped pyramid as we see in 
South India. The central and corner towers are of the North- 
Indian or Sikhara style. An innovation is introduced in Bayon, 
where the towers are capped by four heads facing the four direc- 
tions. The wide ditches surrounding the temples and cities, with 
paved causeways over them, form an important feature of construc- 
tion, and the figures of a series of giants pulling the body of a 
serpent, which form the balustrades of the causeway on its two 
sides, are justly regarded as one of the most ingenious and interest- 
ing architectural devices to be seen anywhere in the world. The 
gallery, referred to above, is in its final shape, a long narrow run- 
ning chamber with vaulted roof, supported by a wall on one side 
and a series of pillars on the other. It has a veranda with a half- 
vaulted roof of lower height supported by columns of smaller 
dimensions, The walls of these galleries are generally covered with 
continuous friezes of bas-reliefs and other sculptures. 

An i'dea of the massive character of these monuments may be 
had from the measurements of Angkor Vat. The moat or ditch 
surrounding the temple and running close to its boundary walls is 
more than 650 ft. wide. This is spanned by a broad stone causeway 


leading to the wall of enclosure which completely surrounds the 
temple and has a total length of two miles and a half. The broad 
paved avenue which runs from the western gateway to the first 
gallery is 520 yds. long and raised 7 ft. above the ground. The 
first gallery measures 265 yds. east to west and 224 from north to 
south, with a total running length of nearly 1000 yds. The central 
tower, on the third or highest gallery, rises to a height of more 
than 210 ft. 

About a mile to the north of Angkor Vat lie the ruins of 
Angkor Thorn, the capital city of Jayavarman VII, formerly be- 
lieved to be that of Yasovarman I. The town was surrounded by 
a high wall made of limonite with a ditch beyond it, 110 yds. wide. 
The ditch has a total length of nearly 8% miles and its sides are 
paved with enormous blocks of stone. The enclosing walls were 
pierced by huge gates which gave access to the city by means of 
five grand avenues each 33 yds. wide and running straight from 
one end to the other. Each gateway consisted of a huge arched 
opening more than 30 ft. high and 15 ft. wide, and surmounted by 
four huge heads placed back to back. The town was square in 
shape, each side measuring about two miles. The grand avenues 
converge to the Temple of Bayon which occupies almost the central 
position of the city and is justly regarded as the masterpiece of 
Kambuja architecture. To the north of the Bayon is a great public 
square, a sort of forum, about 765 yds. long and 165 yds. wide, sur- 
rounded by famous structures such as the Baphuon, the Phimeana- 
kas, the Terrace of Honour etc. The Bayon is the largest temple 
in Angkor Thorn, and though constructed on the same principle as 
Angkor Vat, the arrangements of the galleries are more complex. 
Its lowest gallery is about 165 yds. long from east to west and 109 
yds. from north to south. The central conical tower is crowned 
by four heads, probably of Brahma, and its summit is about 150 ft. 
high from the ground level. Bayon contains some of the finest 
sculptures to be seen in Kambuja. 

These few details would serve to convey an idea of the massive 
character of Kambuja temples. But it is not by their massive 
form alone that they appeal to us. Their fine proportions, the 
general symmetry of the plan, and above all the decorative sculp- 
tures invest them with a peculiar grandeur. 

The sculptures in Kambuja, both bas-reliefs and figures in the 
round, also attained to a high level of excellence. Here, again, 
we find that while the earlier sculptures show a close affinity with 


Indian models, specially Gupta art, new elements are added in 
course of time which give a distinctive character to Kambuja 
sculpture. The peculiar smiling countenance, with half-closed 
eyes, of divine figures, known as the smile of Angkor, has been 
variously interpreted, and opinions differ on its aesthetic values. 
It has been suggested that this unchanging and elusive smile which 
mysteriously reflects the illumination of inward nirvana and ex- 
presses supreme Buddhist beatitude is the most notable contri- 
bution of Khmer art. But this smile of Angkor is not confined 
to Buddhist heads alone, as is generally supposed. It appears in 
Brahmanical images and should, therefore, be regarded as a divine 
expression rather than anything peculiarly Buddhist. Although 
the figures often show traces of Khmer physiognomy, some of the 
best figures, exhibiting plastic quality of a high order, are marked 
by the purity of Aryan profile. 

The bas-reliefs which adorn the temples of Kambuja form the 
most important class of Kambuja sculpture. The earlier speci- 
mens show the figures in fairly high relief like those of Java and 
India, but gradually the depth of the relief is decreased till the 
figures are merely incised or scratched on the surface and the 
whole thing looks like a tapestry on stone. But subject to this 
limitation the bas-relief sculptures show balance, harmony and 
rhythm of a high order. They are marked for their narrative skill 
and cover a wide range of fields embracing almost all phases of 
human and animal lives. The scenes are full of life and move- 
ments and are graceful without being exuberant. The vast lengths 
of galleries covered by these interminable scenes display the deco- 
rative faculties of Kambuja art at their very best, but these are 
duly subordinated to the architecture. 

It is needless to pursue this topic any further. But whether 
we look at the massive temples with elegant proportions or the 
sculptures which adorn their walls, we cannot but pay the greatest 
tribute to their truly classic composition of the highest order. 

If art is an expression of national character and a fair index 
of the culture and civilisation of a people, Kambuja easily takes 
the leading position among the Indian colonies in Indo-China and 
constitutes an immortal landmark and the greatest living testimony 
to the splendour of the civilisation of which it is a product. Un- 
til recently no one doubted the Indian origin of this art, but lately a 
school of critics has sought to establish that this developed phase 
of Kambuja architecture and sculpture is of purely Khmer origin 


and is not indebted to India in any way. They concede the purely 
Indian origin of the earlier structures but hold that the noble monu- 
ments of the llth and 12th century do not owe anything to Indian 
influence or inspiration, but were original inventions of the local 
artists. This is not the place to enter into a detailed discussion of 
this question, but according to a more rational view the Kambuja 
architecture followed a regular course of development from the 
purely Indian type with which it started. It underwent a process 
of evolution such as we notice even in different parts of India itself, 
in different ages, and while the local genius and environments 
added new conceptions of beauty and principles of construction, it 
is as unreasonable to ignore Indian influence upon the monuments 
of Kambuja as to dissociate the culture and civilisation of Kam- 
buja in other spheres from those of India. If we study for example 
the palaeography and iconography we find the same phenomena. 
The earlier alphabets and images are hardly distinguishable from 
those of India, but gradually both undergo slow and steady trans- 
formation. The fully developed Kambuja script of the twelfth 
century A,D. shows an altogether different aspect, and it would be 
difficult to regard it as Indian unless one studies the process of 
evolution. Similarly we find new ieonographic features, new 
names of divinities, and even new conceptions of religion, not met 
with in India. The same has been the case al^o with art and archi- 
tecture. And this is only what we could expect in a living society. 
It is to be noted that the Indian colonisation in the Far East was 
not an imperialism in any form, political or economic. It trans- 
fused new blood, in the shape of the cultural heritage of India, to 
create new life and spirit on alien soils. It transformed the weaker 
and the more backward by fresh vitality, and so long as this life- 
giving force was there the people were quickened by new impulses 
and did not merely imitate but developed healthy lives of their 
own on the foundations well and truly laid by the Indians. What 
the Indian element meant in their life and civilisation is best seen 
when this perennial fountain-source ceased to flow. In proportion 
to the lack of fresh vitalising forces from India the culture and 
civilisation of Kambuja showed signs of decline, and then came the 
inevitable end. It is not perhaps a mere coincidence that the two 
Indian colonies of Champa and Kambuja were overwhelmed by two 
branches of the Thais in the 13th century AJD. when India herself 
lay prostrate under the foreign invaders. The same phenomena 
are also noticed in Java. This sudden collapse of the culture and 
civilisation in these Indian colonies at the very moment when India 
herself lost her independence and was submerged in darkness con- 


stitutes the most important testimony to the influence she exercis- 
ed over their growth and development, 


We do not propose to trace the history of Kambuja beyond the 
twelfth century A.D. but would merely complete this sketch by a 
very brief outline of the subsequent events. 

Jayavarman VII was succeeded by Indravarman II. The latter 
died in 1243 A.D. (No. 186). 

The next king known to us 13 Jayavarman VIII who abdicated 
the throne in 1295-96 AJD. and received the posthumous name 
Paramesvara. It is difficult to say whether he was the immediate 
successor of Indravarman. In that case he would have reigned 
for 52 years. 37a This is not unlikely, for we know that he ruled up 
to ripe old age. For it is said in the Bantay Srei Ins. of his son- 
in-law and successor that the earth, suffering from the troubles of 
the reign of an old man, was now freed from them by the vigilance 
of a young king. 38 But nevertheless Kambuja was powerful enough 
to refuse the payment of tribute asked for by the Mongol Emperors 
of China. 39 

The abdication of Jayavarman VIII was followed by a struggle 
for succession. The old king designated his son-in-law Srlndra- 
varman to be his successor, but his son made an attempt to seize 
the throne. Srindravarman mutilated and imprisoned his brother- 
in-law and ascended the throne in 1296. 40 

Whether the young king was able to remove "the thorns and 
brambles which had grown up 77 during the preceding reign to 
use his own expression it is difficult to say. During his reign 
a Chinese embassy came to Angkor Thorn (1296 A.D.) and Cheu 

37a. In describing the succession of kings after Jayavarman VII I have 
followed Finot's interpretation of the Ins. No. 186. It appears to me, 
however, that there was possibly a king named 6ri Srindrajayavarman after 
Jayavarman VIII, who was succeeded by his son-in-law Indravarman in 1295 
or 1296 A.D. There would thus be two reigns between 1243 and 1296 A.D. 

38. BEFEO. XXXV, p. 395. 

39. BEFEO. H, p. 131. 

40. This is reported by Cheu Ta-kuan; cf. also Ins. No. 185, v. 41. 


Ta-kuan who accompanied it wrote the famous account of the 
manners and customs of Kambuja. 41 

Srindravarman abdicated the throne and was succeeded in 1308 
(No. 185) by Yuvaraja Siindrajayavarman who was related to him 
. 186 vv, 47-8). 

Only another king Jayavarma-Paramesvara is known from the 
Kambuja inscriptions (No. 187). Probably he was the successor 
of Srmdrajayavarman and ascended the throne in 1327 A.D. 42 He 
is the last king known from inscriptions. The subsequent history 
of Kambuja can only be gathered from the Khmer chronicles 
which were composed at a considerably late period and cannot be 
regarded as trustworthy. 

It will thus be seen that our knowledge of the history of 
Kambuja after the death of Jayavarman VII is very poor, and we 
hardly know anything reliable from the fourteenth century on- 

But we can trace the general course of events which led to the 
decline and downfall of the Kambuja Empire. It appears that 
the Thais in tho northern and western parts of the empire were 
organised under able military leaders and openly broke into revolt 
in Siam early in the thirteenth century A,D. A Thai chief Indra- 
ditya founded an independent kingdom with Sukhodaya as capital 
some time about the middle of the thirteenth century A.D. After 
the conquest of the Thai principality of Yunnan by the great 
Mongol chief Kublai Khan in 1254, the newly founded Thai 
kingdom of Sukhodaya received a tremendous wave of Thai immi- 
grants who fled from Yunnan. Ram Kahmheng, the famous Thai 
king of Sukhodaya towards the close of the 13th century, was a great 
conqueror. He carried his arms to Lower Burma on the west and 
to the heart of Kambuja on the east. Cheu Ta-kuan, who visited 
Kambuja shortly after, mentions that in the recent wars with the 
Siamese the region round Angkor was utterly devastated. But 
it is clear from the Chinese memoir that Kambuja was still a 
mighty kingdom and that Ram Kahmheng's invasion was more 
of the nature of a predatory raid than a regular conquest. The 

41. The account has been translated by Remusat in Nouveaux Melanges 
Asiatiques (18290 pp. 77 ff, and by Pelliot in BEFEO. II, pp. 123-177 (for 
some corrections cf. BEFEO. XVHI (4) No. 9.). 

42. BEFEO. XXVm, pp. 145-6. 


Thai kingdom of Sukhodaya came to an end soon after, and a new 
Thai dynasty which founded the kingdom of Ayuthia about 1350 
A.D. soon made itself the master of the whole of Siam and Laos. 
To the east of Kambuja the Annamites gradually conquered nearly 
the whole of the kingdom of Champa by the fifteenth century. 
Kambuja was now hard pressed by these two important Thai 
powers on two sides who steadily ^encroached upon its territory. 
This simultaneous pressure from the two flanks proved the ruin of 
Kambuja. Its weak and helpless rulers tried to save themselves 
by playing off their two 'powerful enemies one against the other, 
but with disastrous consequences to themselves. For centuries 
Kambuja remained the victim of her two pitiless aggressive neigh- 
bours. At last shorn of power and prestige, and reduced to a 
petty state, Ang Duong, the king of Kambuja, in A.D. 1854, threw 
himself under the protection of the French, and thus the once 
mighty kingdom of Kambuja became, as it still is, a petty French 


The following special abbreviations have been used; 

A ft. Aymonier-Le Cambodge, 3 Vols. (Paris, 1900-1903). 
*B Bulletin de VEcole Frangaise d'Extr erne -Orient. 
C=Inscriptions du Cambodge by G. Coedes (Hanoi, 1937). 

Coi=Inscriptions Sanscrites du Cambodge by M. Earth and A. Bergaigne 
(Paris 1885), 


K, followed by a numerical figure, denotes the serial number of the 
inscription in the 'Listes generates des inscriptions et des Monu- 
ments' compiled by G. Coedes and H. Parmentier (Hanoi 1923) and 
in the 'Supplement' published as Appendix to *C' 

LmL. de Lajonquiere Inventairc descriplif des monuments du Cambodge, 
3 Vols. Paris 1902-11. 

Loc Locality. 

Prrr Province, 

'Ed.* denotes that the inscription has been properly edited, and 'R.' 
indicates that it is merely referred to, sometimes with a short summary, 
in the Journal or the Book named after them. 
All the dates are in 6aka Era. 

No. 1. Prasat Ak Yom Door Pillar Ins. D. 531. 

R.B. XXXIII, p. 530, K. 749. Dt. Puok. 
The most ancient Ins. employing Arabic Numerals Foundation 

of the temple (dedicated to Gambhiresvara) by Mratan Kirtigana 
The temple is interred in Baray (western), 

No. 2, Ankor Borei Stone Ins. D. 533. 

R.B. XXXV, p. 491. K. 600. Dt. Prei Krabas. 

It war. found in the ruins of the village where once stood a 

fortified town which was formerly identified with Vyadhapura. 

No. 3. Vat Vihar Tran Pillar Ins, D. 535. 

R.B. XXXV, pp. 36, 142. K. 748, Pr. Kondal Stun. 
The temple contains a linga and an image of Siva with 
Par vat i seated on his knee (p. 122, pi. XVTLA) 

No. 4. Bayan temple Stele Ins. D. 546. 

Ed. Co. No. V, p. 31, K. 13. Dt, Treang 


No. 5. Phnom Bantay Nan Ins. of Bhavavarman (I or II) 

Ed. Co. No. Ill, p. 26, R.A. II. 306. K. 213. Dt, Battambang. 
Installation of a linga by the king. 

No. 6. Prah Vihar Stele Ins. of Bhavavarman (I or II) 
Ed. C. 3. K. 733, Dt. Kompon Len. 

Endowment by poet Vidyapuspa, master of the Pasupata sect, 
and versed in various sastras, viz. grammar, Vaisesika and logic. 

No. 7. Ponhear Hor Pillar Ins. of Bhavavarman (I or II) 
Ed. Co. No. II, p. 21, K. 21. Dt. Tran. 

Installation of Linga, Durga, ambhu-Visnu, Visnu Trailokya- 

No. 8. Han Chey Stele Ins. of Bhavavarman (I or II) . 
Ed. Co. No. 1, p. 8, K. 81. 
Eulogy of the king and donations of a linga by an official. 

No. 9. Veal Kantel Ins.. of Bhavavarman I. 

Ed. Co. No. IV, p. 28, R.A. II, 180, K. 359. 
Installation of an image of Tribhuvanesvara by Brahman Soma- 
sarman, husband of the daughter of Viravarman, who was sister 
of Bhavavarman and mother of Hiranyavarman. Donation of 
complete texts of Mahabharata, Ramayana and Purana. 

No. 10. Thma Kre Ins. of Citrasena. 
Ed. B. Ill, p. 212, -K. 122 
Loc. on the Mekong south of Sambor. 

One verse only Installation of a &iva-lmga by Citrasena at the 
command (or with the approval) of his parent (Matd-pitrorrr 
anujnayai) . There are 2 replicas of this inscription (1) Cruoy 
Amphil. R.B. IV, p. 739. K. 116 (2) Tham Pet Thong (Rajaxiraa 
of Siom). RB. XXII, p. 92, K. 514. 

No. 11. Chan Nakhon (Phu Lokhon) Ins. of Mahcndravarman 
Ed. B. III. 442. R.L. II. 73. K. 363, Dt. Bassak. 
Replicas (1-2) Khan Thevada. At the confluence of the Mun and 
the Mekong rivers. 

(3) Tham Prasat 

(4) Vat Xumphon at Muang Surin 
B. XXII, pp. 57-59. 

(5) Keng Tana (on the Mun) 
B. XXII, p. 385. 

For complete text cf. B. XXII, p. 58. 

Installation of a liiiga on the top of the hill (or bull), as a sign 
of victory by Citrasena (grandson of Sarvabhauma, son of 
Viravarman, youngest brother of Bhavavarman) who took the 
coronation name of Mahendravarman. 

No. 12. Prasat Pram Loven Stele Ins, of Isanavarman. 
R.A.I, p. 140. L. Ill, p. 478, K, 7. Dt. Sadec. 
Installation of a divine image by the king. 


No, 13. Sambor Prei Kuk Ins. of Isanavarman. 
R.B. XIII (1), p. 27, Dt. Kompon Svay. 

Records a pious foundation by the queen Sakara-manjari 
(BCAI, 1912, p. 188). 

No. 14. Sambor Prei Kuk Ins. of Isanavarman. 
R.B. XIH (H), p. 28. 
Installation of an image of Prahantesvara by the king. 

No. 15. Sambor Prei Kuk Ins. of Isanavarman. 

R.B. XIII (1), p. 28; BCAI, 1912, p. 187. 

Installation of the images of Sarasvati, Nrttesvara and Nandin, 

and donations by the queen. 

No. 16. Sambor Prei Kuk Ins. of Isanavarman. 

RJB, XIII (1), p. 28; BCAI, 1912, p. 188, Dt. Kampon Svay. 
Mentions Isanavarman and his relation with India, cf. the 

No. 17. Sambor Prei Kuk Ins. 

R.B. XIII (1), p. 28, Dt. Kampon Svay 
Mentions Mahendravarman and ibaiiavarman, 

No. 18. Ang Pu (Vat Pu) Stele Ins. of Isanavarman. 
Ed Co. No. VIII, p. 47. K. 22, Dt. Tran. 

Installation of an image and a linga of Siva-Visnu by Muni 
Isanadatta . 

No. 19. Svai Chno Stele Ins. of Isanavarman. 

d. Co. No. VII, p. 44. K. 80. Dt. Kandal. 
Foundation of an dsrama by Arya Vidyadeva. 

No, 19 A. Vat Sabab Ins. of Isanavarman. 

Ed. B. XXIV, p. 353. Loc. South-east of Chantabun. 

No, 20. Sambor Prei Kuk. Pillar Ins. of Isanavarman D. 549. 
Ed. B. XXVIII, p. 44. K. 604. Dt. Kampon Svay. 
Homage to Sri Kadambesvara Eulogy of king Sri Isanavarman 
one of his servants versed in Grammar, Vaisesika, Nyaya 
Samlksa and Buddhism erection of a Linga priest, a Pasupata 
Brahmana king referred to as lord of Tatandarapura (?), 

No. 21. Prah vihar Kuk (or Vat Chakret) Stele Ins. cf Isanavarman 
D. 549. 

Ed. Co. No. VI, p. 38. 

R.A.I., 237; L.I., 58; K. 60; Dt. Prei Ven (Ba Phnom) . 
Refers to king Isanavarman and records erection of an image 
of Siva-Visnu by the vassal ruler of Tamrapura. 

No. 22. Vat Kdei Ah (Ang Chumnik) Stele Ins. D, 550. 

Ed. Co. No. IX, p. 51; R.A.I., p. 241; L.I., p. 54; K. 54. 

Dt. Prei Ven (Ba Phnom) . 

Refers to the restoration (?) of a Siva-Zinga by Acarya Vidya- 

vinaya and also erection of two lingos, Refers to Jayavarman 

in V. 7-B, which is evidently a later addition. 



No. 23. Phnom Bayan Ins. of Bhavavarman II. 
Ed. C. 252; K. 483. 

Very fragmentary. Donation to Utpannesvara. Mentions 

No. 24. Phnom Penh Stele Ins, of Bhavavarman II, D. 561. 

Ed. B. IV, p. 691. R.B.I., p. 161; XV (2), p. 26. L.I., p. 82; K. 79. 
Probably belonged originally to the Dv. of Takeo and is now 
at Eccle Franchise Installation of Devlcaturbhuja by the king, 
and donations. 

No. 25. Tuol Kok Prah Stele Ins. of king Jayavarman I, D. 579. 
Ed. B. XVIII (10), p. 15; K. 493; 

Loc, Paddyfield, east of Tuol Kok Prah, Kompong Rursei, 
Dt. Prei Ven. 

Foundation of the temple of God Amratakesvara by Jnanacandra, 
minister of Jayavarman I, in 579. 

No. 26. Vat Prei Var Stele Ins. of king Jayavarman I, D. 587. 

Ed. Co. No. X, p. 60, R.A.I., p. 248; L.I. p. 51; K. 49; Dv. 

Prei Ven. 

Donation by king Jayavarman I to two Buddhist monks. 

No. 27. Kdei An (Ang Chumnik) Stele Ins. of king Jayavarman I 
D. 589 (past) -667 A.D. (C. 12). 

Ed. Co. No. XI, p. 64; A,L, p. 243; L.I. p. 54; K. 53; Dv . 
Prei Ven. 

Installation of a lingo, and the endowment of a temple dedicated 
to Siva Vijayesvara in the town of Adhyapura by Simhadatta, 
physician and hereditary governor Refers to kings Rudravar- 
man, Bhavav , Mahendrav and Isanav . Four generations of 
donors, one of whom was sent on an embassy to Champa, 
another, a poet and a minister. 

No, 28. Vat Prei Var Stone Slab Ins.; D. 589 (past). 

Ed. Co. No. XII, p. 73; A. I., p. 51; K. 50; Dv. Prei Ven. 
Installation of a Uhga of Visnu-Isa (Harihara?; 

No. 29. Vat Kirivon Stone Ins. D. 592. 

B.B, XXX, pp. 526, 583. K. 666; Dt. Tralac. 
Installation of a lihga. 

No. 30. Tuol Prah That Pillar Ins. of Jayavarman I, D. 595. 
Ed. C. 12; K. 762. 
Loc. Khand of Kancriec. 
Installation of a linga (Kedaresvara) by Rajasabhadhipati. 

No. 30A. Prah Kuha Luon Ins. of Jayavarman, D. 596; R.C. 13. 

No. 31. Vat Baray Stele Ins. D. 598. 

Ed. Co. No. Xm, p. 75; K. 140; Dv. Baray (Kampong Savay). 

Installation of an image of {Sambhu Ref. to gankara- 

No. 32. Tan Kran Stele Ins. of Jayavarman I. 

Ed. C. 7; K. 725; Dt. Con Prei (Kompoong Chan), 


Invocation to Pingalesa genealogy of a family of royal officials 
(designations of various offices) Eulogy of the king Refers to 
Kafichipura Governor of Dhruvapura (full of dense forest) 
inhabited by savages (V. XIV), Refers to Sresthapura, 

No, 33. Vat Phu Stele Ins. of Jayavarman I. 

13d. B. H, p. 235; K. 367; Dt. Bassak, Laos. 

No. 34. Sambor Ins. of Jayavarman I. 

R.A.I., p. 306; K. 131; Dt. Krace. 

No. 35. Snay Pol Stele Ins. (6th Cent. S). 
Ed. B. XV (2), p. 21; K. 66. 

Records gifts to Bhagavati by Salgramasvami and Aditya Svamt 
(figures of 6alagrama and sun, after the names of donors). 

No. 36. Prasat Prah That Ins. (6th Cent. S') . 

Ed. B. XI, p. 393; K. 109; Dt. Thbon Khmum 

Gift of a Ms. (pustaka) of Sambhava (Sarnbhava -par vain of 

Mahabharata) . 

No. 37. Vat Thlen Stele Ins. (6th Cent. 3). 

R.A.I., p. 146; L. III. p. 479; K. 1. Dt. Chaudoc. 

Donations of the Governors of Jesthapura and Bhavapura to God 

Sarikaranaray ana . 

No. 38. Prasat Pram Loven Stele Ins. (6th Cent. 6). 

R.A.I., p. 139; L. III. p. 478; K. 6. Dt. Sadec (Cochin-China) . 
Recoids pious works by a priest 6ri Puspavatasvaml, donations 
to God Mulasthana by a high official Sucidata, Yayamana 
Vrah Kamratan (high dignitary) 6ri Puskaraksa (the ancestor 
of Yasovarman (?). 

No. 39. Prasat Pram Loven Stele Ins. (6th Cont. ). 

R.A.I., p. 140; L. HI. p. 478; K. 8. Dt. Sadec. 

Pious donations by a chief Krsnadatta to god 3rf Amratakesvara. 

No. 40. Camnom Stele Ins. (6th Cent. ). 

R.A.I., pp. 195-96; JA. 1883 (1), p. 450. L.I., p. 16. K. 30. 
Dt. Prei Krabas. 

Installation of god Yajnapatisvara (Harihara?) by Krsnamurti; 
construction of temple and donations by his relations. 

No. 41. Nan Khmau Stele Ins. (6th Cent. S). 

R.A.I., p. 183; L.I., p. 31. K. 37: Dt. Bati. 
Donations to god Vrah Yama. 

No. 42. Vat Tnot Stele Ins. (6th Cent. g). 

R.A.I., p, 182 JA., 1883 (1), p. 449, L.I., p. 32. K. 38, 
Dt. Bati, 

Donations to gods Sri Ganapati, Vrah Svayambhu (Brahma) 
Refers to Vikramapura (old town near Bati) mentions Jaya- 
varman (king Jayavarman I ? ) . 

No. 43. Vat Pret Sva Stele Ins. (6th Cent. 3) . 

R.A.I., p, 182. L.I, p. 45. K. 41. Dt. Bati, 
Donations to god Pihgales vara ($iva ? ) . 


No. 44. Phnom Ngouk Stele Ins. (6th Cent. S). 

R.A.I., p. 154. L.I.,p. 49. K. 46. Dt. Kampot. 
Donations to god Utpannesvara (Utpalesvara ? Siva?). 

No, 45. Sambor Ins. (6th Cent. 6). 

R.B. IV, p. 742, f.n. K. 133. Dt. Kraceh. 
Donations of a high official called Maha-nauvahaka. 

No. 46. Ba Dom Stele Ins. (6th cent. 3). 

R.B. Ill, p. 369, Lu II, p. 64. K. 360. Dt. Stun Tren (Laos). 
Construction of a brick temple of iva and a bhaktaadla and a 

No. 47. Vihar Thorn Trisula Ins. 

Ed. B. XX, pp. 6-7. K. 520. Dt. Kompon Sien. 

Records that a tooth of an octogenerian was deposited under a 

Trisula erected by him. 

No. 48. Prah That Kvan Pir Pillar Ins. D. 638. 

Ed. B. IV, p. P-75; R.L.I., p. 185. K. 121. Dt. Kraceh. 
Installation of god Puskaresa by Puskara. As suggested by 
Pelliot (B. IV. p. 214) this Pusakara may be identified with 
Puskaraksa, prince of Aninditapura and king of Sambhupura 
(Co. p. 356-7). 

No. 49. Lobok Srot Pillar Ins. of Jayavarman, D. 703. 

Ed, B.V., p. 419. R.B. XXVIII, p. 119, K. 134. Dt. Kraceh. 
The king is referred to as Brahma -ksatra-vamsa 
Vaisnava foundations. 

No, 50. Vat Tasar Moroy Stele Ins. D, 725. 

R.A.I., p. 305; L.I., p. 187. K. 124. Dt. Kraceh. 
Donation of the queen Jyestharya to Siva mentions Jayendra, 
queen Nrpendradevi and the king (who has gone to) Srindraloka 
Donation (probably of the same queen) to God Srimadamrataka. 

No. 51. Prasat Prei Kmen Ins. D. 782. 
R.B. XXXin, p. 1137. 
Refers to Jayavarman III. 

No. 52. Prasat Kok Po Ins. of Jayavarman HI. 
R.A.I., p. 384. K. 256. Dt. Siem Rap. 
Donations by king Visnuloka to god Pundarikaksa 6vetadvipa. 

No. 53. Prasat Cak (Angkor) Ins. of Jayavarman III. D. 791. 
Ed B. XXVUI, p. 115, K. 521. 
The 16th regnal year of the king Installation of Saka Brahmana, 

No. 54. Prah Ko Stele Ins. of Indravarman, D. 801. 
Ed. C. p. 19. K. 713. Dt. Siem Rap. 

Indravarman became king in 799, 6. Instals three images in 801. 
Nos. 54-57 and 59 are nearly identical, cf. C. 17. 

No. 55. Bakong Door Pillar Ins. of Indravarman, D, 801(7). 
R.Co. No. XXXVII. p. 310. K. 304-308. 
Five copies of the inscription. 


No. 56. Bako Stele Ins. of Indravarman. D. 801. 

Ed. Co. No. XXXVI. p. 297. K. 310-13. also 315-22 (of the 
same purport), (six copies). Dt. Siem Rap. 
Genealogy of the king and his donations. 

No. 57. Bakong Stele Ins. of Indravarman, D. 303. 
Ed. C. 31. K, 826. Dt. Sutnikom. 

Foundation of Linga Indresvara first 22 stanzas identical with 
Prah Ko (No. 54). 

No, 58. Prasat Kandol Dom Door Pillar Ins. of Ind> avarman, D. 801. 
Ed. C. 37, K. 809. 

Mentions Indravarman and his guru ivasorna who learnt Sastras 
from Sankaracarya. 

No. 59. Bayang Stele Ins. of Indravarman I. 

Ed. Co. No. XXXVin, p. 312. K. 14. Dt. Tran 

Foundation of a temple and two monasteries by the king at 


No. 60. Prah Bat Stele Ins. of Yasovarman, D. 813. 

Ed. Co. No. XLIV, p. 355. K. 95. Dt. Con Prei. 

Genealogy of the king Regulations of the monasteries Pious 

foundations and eulogy of the king. 

There are ten replicas of this inscription (cf. Co. Nos. XLV 

LIV. pp. 376-390) which contain identical verses except one 

which refers to the particular divinity for whom the Ins. is meant. 

No. 61. Loley Stele Ins. of Yasovarman. 
Ed. Co. No. LV. p. 391. 
Royal genealogy as in No. 60, but the eulogy is different. 

No. 62. (A-F), Six Thnal Baray Stele Ins. of Yasovarman, 
Ed. Co. Nos. LVI-LX, pp. 420-525; B. XXXII, p. 85 
62A Co. LVI; 62 B-Co. LVII; 62CrzCo. LVTII; 62DrzCo. LlX; 
62E=Co. LX. 62F. Ed. B. XXXII, p. 85. 
These six inss. are nearly identical. 

No. 63 Prasat Komnap Ins. of Yasovarman. 
Ed. B. XXXII. 88. K. 701. 

Mostly identical with 62A. Genealogy identical with 61. 
Regulations of a Vishnuite Asrama. 

No. 63 A. Tep Pranam Ins. of Yasovarman. 

Ed. JA, 1908 (1), p. 203; 1908 (2), p. 253. K. 290. 
Nearly identical with Nos. 62A and 63. 

No. 64. Bako Door Pillar Ins. D. 813. 

R.A. II. p. 444. K. 314. Dt. Siem Rap. 
Donations of Isvaravarman to Isvarasrama. 

No. 65. Prah Ko Stele Ins. of Yasovarman, D. 815. 
Ed. C. 28. K. 713. (See No. 54). 
Donation of Yasovarman to Parame6vara. 


No, 66, Loley Door Pillar Ins. of Yasovarman, D. 815. 

Ed. Co. XXXIX-XLH (pp. 324-331). K. 324; 327; 330; 331. 

No, 67. Phnom Prah Vihara Pillar Ins. D. 815 

Ed. Co. LXI, p. 525; K. 382. Dt. Mlu Prei. 

Mentions king Jayavarman II with 724 as the date of accession 

and eulogises Sivasakti. 

No. 68. Phnom Dei Door Pillar Ins. of Yasovarman. D. 815(?). 
Ed. B. XVIII (9), p. 13. 

Temple on the summit of ri-Purandara-parvata dedicated 
to Harihara. 

No. 69. Phnom Sandak Stele Ins. of Yasovarman, D. 817. 
Ed. Co. XLIH, p. 331. K. 190. 

Eulogy of the king and previous kings, particularly Jayavar- 
man II. 

Foundations of Somasiva, adhydpaka, nominated by the king. 
The same date is given in Prasat Prei Kemen Ins. B. XXXIII, 
p. 1137. 

No. 70. Phimanakas Door Pillar Ins. of Yasovarman, D, 832. - 
Ed. Co. No. LXII, p. 545. K. 291. 
Loc. Angkor Thorn. 

Construction of a temple of Visnu Records the date of king's 
death (?) 

Nos. 71-72. Two Angkor Thorn Inss. of Yasovarman. 
Ed. B. XXV, pp. 305-9. K. 491, 576. 
Donations by the king's uncle Jri Samaravikrama. 

No. 73. Phnom Bay an Ins, of Yasovarman. 
Ed. C 256. K. 853. 

Eulogy of Amarabhava, highly esteemeed by the king and 
appointed by Indravarman as chief of Indrasrama. 

No. 74. Vat Thipedi Door Pillar Ins. of Isanavarman II. D. 832. 

Ed. Melange S. Levi, p. 213. R.A. II., p. 379. K. 253. 
Dt. Siem Rap. 

It contains two different Inss. A and B. A gives eulogy of Yaso- 
varman and his two sons, and records the erection of the temple, 
in 83?, by Sikhasiva, Marginal Text in Khmer records some 
donations of another person in 834. B. refers to Suryavarman 
and commemorates the restoration, in 927, of a Hnga, consecrated 
95 years ago Sikhasiva, predecessor of a certain Krtlndrapantfita 
of whom it gives the genealogy, giving a new instance of 
succession, in female line, of hotars of different kings. (The 
date 832 may not belong to the reign of Isanavarman II. Cf. JGIS. 
HI, p. 65) . 

No. 75. Vihar Kuk (Vat Cakret) Stele Ins. of Harsavarman I. D. 834 (?). 
Ed. Co. No. LXm, p. 551. R. JGIS. Ill, p. 65. K. 61. Dt. Prei 
Donation of the king to Adrivyfidhapuresa (Siva). 


No. 76. Prasat Thorn (Koh Ker ) Ins. of Jayavarman IV. D. 843. 
Ed. B. XXXI, p. 13. Co. No. LXIV, p. 555. K. 682. 

No. 77. Two Koh Ker Inss, D. 843. 

Ed. B. XXXI, p. 15. K. 682. 

No. 78. Tuol Pel Ins. of Harsavarman (?) D. 844. 

Ed. B. XXXI, p. 17. R.A. I. p. 443. JGIS, III. p. 65. K. 164. 
Dt. Sron. 

The name of the king is doubtful, but, if correct, it is the Oiil^ 
Ins. datable in his reign. 

No. 79. Con Ah Pillar Ins. of Jayavarman IV. D. 844. 

R.B. XXXI, p. 16; A.I. p. 292. K. *>9. Dt. Thbon Khmum. 
Order of the king to Prthivindravarman Installation of gods 
Tribhuvanaikanatha and Campesvara (Krsna ? ) and donations. 

No. 80. Prasat Neang Khmau Ins. of Jayavarman IV. D. 850. 
R.A. I, p. 183. K. 35. Dt. Bati. 
It gives the year 850 as the date of the accession of the king. 

No. 81. Koh Ker Pillar Ins. of Jayavarman IV. D, 851 (?), 852, 854. 
Ed. Co. No. LXIV, p. 555. K. 184, 186, 187, 188. 

No. 82. Prasat Andon Ins. of Jayavarman IV. 
Ed. C. 61, K. 675. 

No. 83. Prasat Damrei Ins. of Jayavarman IV. 
Ed. C. 56. K. 677. 

No. 83A. Prasat Kok (wrongly described as Prasat Preah Dak) Ins. of 
Jayavarman IV. 

R.A. II. p. 419. K. 339. Dt, Siem Rap. 

Invocation of the three Buddhist Ratnas genealogy of the king 
from Jayavarman II Conquest of Champa by Jayavarman IV. 

No. 84, Phnom Bayan Ins. of Harsavarman II. D. 863. 
d. C. 260. K. 854. Dt. Tonlap. 

Invocation to Utpannakesvara followed by the eulogy of Jaya- 
varman IV who appears, from some expressions, to have usurped 
the throne. The date is one year earlier than that generally 
assumed for the accession of the king. 

No. 85. Vat Kdei Car Stele Ins. of Harsavarman II. D. 864. 

R.A. I. p. 372; B. XV (2), p. 25. K. 157. Dt. Kompon Svay 
Gives the date of the accession of the king. 

No. 86. Trapan Samliot Stele Ins. of Rajendravarman, D. 866. 
R.A. I. p. 165. K. 19. Dt. Tran. 
Donations. Year of accession. 

No. 87. Two Prah Put Lo Rock Inss. D. 869. 

Ed. JA. 1914 (1), pp. 638, 644. R.A. I, p. 426. K. 173-4. 
Religious aphorisms; eulogy of their authors who were ascetics. 

No. 88. Prasat Pram Door Pillar Ins, of Rajendravarman D. 869 (Febru- 
ary, 948 A.D.). 


Ed. B. XIII (6), p. 17. K. 180. Dt, Kompon Svay. 
Eulogy of king Jayavarman IV and his two sons. Installation of 
two lingas by Rudracarya, the teacher of the king, and a pupil 
of Sivasoma, the famous guru of Indravarmari, 

No. 89. Baksei Camkron Door Pillar Ins. of Rajendravarman, D. 869. 

Ed. JA. 1909 (1), p. 467. R.A. IE, p. 80. K. 286. Loc. Mt, 


It gives the mythical story of the foundation of Kambuja by 

JJisi Kambu and refers to the kings 6rutavarman, Rudravarraan, 

Jayavarman II and his successors. 

No. 89A. Mebon Ins. of Rajendravarman, D. 874. 

Ed. B. XXV, p. 309. Loc. Near Angkor Thorn. 
Genealogy and eulogy of Rajendravarman. 

No. 90. Thvar Kdei Ins. of Rajendravarman. D. 874 (or 871?) and 879. 
R.A. I, p. 444. K. 165. Dt. Sron. 

Queen Mahendradevi informs king Rajendravarman of the terri- 
tories enjoyed by her ancestors in Dvaravati, Sahakara, and 
other lands. Donations to Campesvara. Invocation to Visnu, 
called Vasudeva, Hari, Narayana, and Madhvari, identified with 

No. 91. Phnom Sandak Stele Ins. D, 878. 

R.A. I, p. 393. K. 102. Dt. Cikren. 

Royal donations to Sivapura (Phnom Sandak) . 

No. 92. Bat Cum Ins. of Rajendravarman, D. 882. 

Ed. JA. 1908 (2), p. 213. R.A. Ill, p. 11. K. 266-8. Dt. Siem Rap. 
Eulogy of Rajendravarman who embellished Yasodharapurl, 
deserted for a long time, and destroyed Champa and other 
foreign kingdoms. 

Mahayana Buddhist Divinities Eulogy of the Buddhist minister 
Kavlndrarimthana and his pious foundations. 

No. 93. Pre Rup Stele Ins. of Rajendravarman, D. 883. 
Ed. C. 73. K. 806. Dt. Siem Rap. 
Genealogy of the king. Conquest of Champa by the king. 

No. 94, Phncm Trap Ins. D. 875, 882, 884. 

R.A. I. p. 322. K. 94. Dt. Con Prei. 

Installation of the images of Aja (882) and of Upendra (884). 

Arrival of Bhadrayogisvara in 875. 

No. 95, Neak Ta Carek Ins. D. 884. 

R.A. I, p. 384. K. 181. Dt. Cikren. 

Judgment of the king against Vlrabhaktigarjita, chief of Vlra- 
pura, who had removed the boundary and reaped the corn of a 
field which was granted to another person. The chief was fined 
10 ounces of gold. His younger brother, who ordered the reap- 
ing of corn, and another who instigated the crime were given 102 
stripes on their backs. 


No, 96. Don Tri Ins. of Rajendravarman, D. 888. 

R. A. II, p. 283. K. 198 Dt. Battambang. 

Royal order to a number of officers whose names and offices are 

given Buddhist divinities Paramos varaaya-maitrideva. 

No. 97. Bantay Srei Ins. of Jayavarman V, D. 890 
Ed. C. 147. K. 842. 

(Almost a replica of K. 619, B. XXVIII. p. 4t> and ot K. 662, 
B. XXIX, p. 292). 

Foundation of the temple by Yajnavdrohn, the gut u uf the king 
Eulogy of Rajendravarman who conquered ChampaEulogy oi 
the king. 

No. 98. Tuol Kul Ins. D. 890. 

R. JGIS, HI, 65, B. XXXV, p. 493. K. 831, Dt. Mon. . 
Refers to an address presented to I^anavarman IT in 847 aka, 
the only known date of this king and evidence that he actually 
ascended the throne. 

No. 99. Angkor Vat Ins. of Jayavarman V. D, 890. K. 579. 

Phnom Bakhen Ins. of Jayavaiman V. D 890. K. 464. 

Identical hi some parts. 

Ed. B. XXV, p 363. B. XI, p. 3%. 

Gives the date 890 as the commencement of his reign. 

No. 100. Bantay Srei Ins. of Jayavarman V. D. 891. 
Ed. C. 144. K. 570, 

Foundation of the king to Tribhuvaiiamahesvaia (i.e. Bantay 
Srei) . 

No. 101. Bantay Srei Ins. D. 891. 

Ed. Memoires Arch EFEO Nos. 2, 74. K. 571. 

No. 102. Kok Svay Prahm Ins. D. 891. 

Ed. C. 187. K. 848. Dt, Sutnikom. 

Royal order to the yrdmavrddha and purn^apradlt^'ia of Hari- 
haralaya. It proves that Svay Prahm which forms pait of the 
Roluos group was situated in the territory of Hariharalaya. 
This confirms the view of Coedes (B. XXVIII, p. 121) that 
Roluos represents Hariharalaya where Jayavarman II lived 
twice and died, and which was ihe capital of his successors till 
Yasovarman I founded Yasodharapura on the site of Angkor. 

No. 103. Prah Einkosi Ins. of Rajendravarman D. 890 and Jayavarman 
V., D. 892. 

Ed. Co. No. XIV, p. 77; C. 160. K. 262, 263 (K. 668 Replica) . 
Dt. Siem Rap. 

A Refers to Rajendravarman's predecessor, a king of the race 
of Kaundinya who lived in Aninditapura. 

B Eulogy of Jayavarman and the diverse foundations of his 
younger sister Indralaksmi and her husband, the Brahmana 
Divakarabhatta, a native of the bank of the Yamuna (in India), 

No. 104. Kok Rosi Ins. of Jayavarman V. D. 891, 
Ed. B. XXVIII, pp. 113-14. 


No. 105. Basak Stele Ins. of Rajendravarman. 

Ed. B. XV (2), p. 22. K. 70. Dt. Romduol. 
Donation of a chief named Nrpendrayudha, parsvadhara of the 
king to god Vakakakesvara Refers to the installation by the 
king of five images at Angkor on the island of Mebon in the 
centre of Thnal Baray (Eastern) (Yasodhara-tataka). 

No, 106. Basak Stele Ins. 

R,B. XV (2), p. 20, K. 71. 

Religious foundation by Rajakula Mahamantri (minister of 

Rajendravarman) . 

No. 107. Srey Santhor Ins, of Jayavarman V. 

Ed. Revue Archeologique, 1883, pp. 182-192. 

No. 108. Prasat Komphus Ins. of Jayavarman V. D, 894. 
Ed. C. 159. K. 669. 
It is almost a replica of No. 103. 

No. 109. Prasat Nak Buos Ins. of Jayavarman V. D. 896, 
R. Co. 381. K. 343. 
Royal gift to ivapada. 

No. 110, Phnom Bantay Nan Ins. D. 902, 903. 

R.A. II, p. 306 (with commentaries of Kern). K. 214. 
No. 5 above refers to it as Saiva temple under Bhavavarman 
but this inscription refers to Mahayana Buddhist divinities, on 
which Kern has commented 

No. Ill, Prah Einkosi Loss. D. 883, 890, 902, 904, 906, 
R.A. II, pp. 407-410. (cf. No. 103 above). 
Foundation in favour of a monastery called Vidyasrama (883 ), 
Temple of Dwijendrapura (890), Divakarabhatta, priest of this 
temple, receives donations (902) which are continued (904) and 
added to (906). Dwijendrapura (Temple of Pi ah Einkosi) 
inherits parts of the slaves of Vidyasrama. 

No. 112. Prasat Car Ins. of Jayavarman V. D. 901, 916. 
R.A. H, p. 387. K. 257. Dt. Siem Raj 

Installation of various Brahmanical divinities and donations to 
them by Narapativlravarman. 

No. 113, Prasat Trapan Con Ins. 

Ed. B. XXIX, p. 292, Fragmentary; almost identical with No. 114. 

No, 114. Prasat Sek Ta Tuy Ins. of Jayavarman V. 

Ed. B. XXVUI, p. 46. R.B. XXIX. p. 291. f.n.l. K. 617. 
Dt. Cikren. 

Jayavarman's victory in Champa. Reference to Srlparvata in 
the Dak^inapatha Religious acts and donations of the royal 
guru Yajnavaraha. 

No, 115. Angkor Thorn Ins. 9th Cent. 

Ed. B. XXIX, p. 343. JA. Vol. CCXX (1932) p. 50. K. 643, 
Invocations to various forms of Vi$nu and other gods. 


No. 116. Pon Pra Thvar Grotto Ins 9th or 10th Cent, 
Ed. B. XL 398. K. 172. 
Interesting account of a cave 

No. 117. Prasat Khna Ins. of Udayfidityavarman D. 902, 923, 
Ed. B. XI, p. 400. K. 356. Dt. Mlu Prei. 

It definitely proves the existence of the king and gives his date 
and genealogy showing his relationship with king Jayavar- 
man V. 

No 118. Prasam Thorn Ins. of Udayadityavarman D. 923 
Ed. C. 50. K. 682. Loc. Koh ker. 

Royal order about donations to Prthivmarendra and Vlrendrari- 

No. 119. Prasat Ak Yom Ins. D. 923. 

R.B. XXXIII, p. 531. K. 752. Dt. Puok 

It proves that the 'Western Baray' beneath which the stone 

was interred, was excavated after 923. 

No, 120. Sambor Ins. D. 923. 

Ed. B. XXVIII, p. 142. R.A. I, p 307. K. 125. Dt. Kraceh. 

It definitely locates Sambhu-pura, famous since the sixth cen- 
tury. It may refer to Udayadityavarman I or iSuryavarman. 

No. 121. Stun Crap Ins. of Jayaviravarmadeva D. 925, 

R.B. XXXIV, p. 423, B. XXXI, p, 620 K. 693. Dt. Battambang. 
Application of Brahmaputra claiming a foundation made by his 
ancestor Panditankura Acarya Dharmadhipati before the time 
of Ysovarman. The judgment of the king in favour of the 
claimant There is an image of Yama (Dharmadhipati). 

No. 122 Tuol Prasat Ins. of Jayaviravarman D 925. 

R.A. I, pp. 379-381. B. XXXIV, p. 423; K. 158. Dt. Kampon 


GivQs the date 924 for Jayavarman and 925 for Jayaviravarman 

living in Jayendranagarl Refers to donations made by many 

previous kings. 

No. 123. Prasat Kok Po Inscription of Jayavarman D. 900, 906, (also dates 
901, 926) . 
Ed. B. XXXVII, p. 379. K. 255. 256, 814. 

No. 124. Prah Ko Ins. of Jayaviravarman D. 927. 
Ed. C. 189. K. 717. Dt. Sutnikom, 

Sikhasiva's grandson Vinaya, a Professor, obtained favour from 
the king. It proves that Jayaviravarman was different from 
Suryavarman gives the date of accession of Jayavarman IV 
(850) and Rajendravarman (866). 

No. 125. Prasat Dambauk Khpos Ins. of Jayaviravarman D. 927. 
R.A. I., p. 420; B. XXXIV, p. 423. K. 196. 

No, 126. Roban Romas Ins. of Suryavarman I. D. 923. 

R.B. XXXIV, p. 422. K, 153. Dt. Kompon Svay, 
Donation of SomeSvara Pandita. 


No, 127. Prasat Tapan Run Ins. D. 924 (or 934|). 

R.B. XXXIV, p, 422. K. 705. Dt. Kompon Svay. 

No. 128. Prah Nan Ins. of Suryavarman I. D. 924. 
R.A. I, p. 328. K. 89. 

The king founded Bhadresvarasrama for the gods Liiigapura 
and Lingasodhana. It was consecrated by 6ri Prathivindra 
Panijita of the country of Ramani Donations to dcdryas living 
in the monastery Refers to the hereditary governor of Bhava- 
pura deified statue of Jayavarman II god Jalangesvara, 

No. 129. Tuol Don Srei Ins, of Suryavarman I. D. 924. 

R.B. XXXIV, p. 427; XXXV, p. 493. K. 834. Dt. Baray, 
Kompon Thorn. 

Refeis to a war of 9 years by Suryavarman who became king 
in 924. Gives the history of a family of royal officials since 
the reign of Jayavarman II. 

No, 130. Tep Pranam Ins. of Suryavarman I. D. 927. 
R.A. HI, p. 112. K. 290. 

No. 131. Vat Phu Ins. of Suryavarman I. D. 928 

R.B. XXXIII (531). K. 720. Dt. Bassac. 

No. 132. Prasat Trapan Ins. of Jayaviravarman D. 928. 
Ed. B. XXVin, p, 58. K. 598. Dt. Sutnikom. 
Donation to a temple (Vaisnava) of land in Amnditapura 
genealogy of Pancagavya or Kavindra-pandita who founded the 
temple and whose ancestors served the kin^s from Jayavarman II 
It fixes the location of Amnditapura (p. 61). 

No. 133. Phnom Prah Net Prah Ins. of Jayavivavarman D. 927-29. 

R.A. II, p. 322. B. XXXIV, p. 423. K. 216. Dt. Battnmbang. 
Order of the king (927) foundations at Sivapada or aivapada 
(928); a royal order (king not named) in 929. 

No. 134. Phnom Sanke Kon Ins. of Suryavarman I. D. 928, 929, 

R.A. II, p. 246. B. XXXIV, p. 424. K. 232. Dt. Krabin (Siam) 
Royal order about a donation. 

No. 136. Phimanaka Inscriptions of Suryavarman I, D. 933, 
R.A. II, p. 233. K. 342. Dt. Mlu Prei. 
Royal donations. 

No. 136. Phimanaka Inscriptions of Suryavarman I, D. 933. 
Ed. B. XHI (6), p. 11. K. 292. Loc. Angkor Thorn. 
Eight Inss. on the pillars of Gopura leading to the interior 
of the royal palace at Angkor Thorn, reproducing, in identical 
terms, the formula of oath pronounced by certain officials of 
the court of the king. 
Two other replicas of the same. 
R.B. XIII (6), p. 12. 

No. 137. Inscriptions of Bantay Srei D. 933. 

Ed. Memoir Arch. EFEO, I. Nos. 3, 4, p. 77. K, 569, 572. 
For other Ins. K. 573-575, cf, Ibid, Nos. 8, 7, 6, 


No. 138. Phnom Cisor Ins. of Suryavarman I. D. 937, 939, 941. 
R.A. I, pp. 191-92. K. 33, 31. Dt. Bati. 

Foundation of monasteries called Yogendralaya and Yogendra- 
pura god Vrddhesvara. 

No. 139. Lopburi (Siam) Ins. of Suryavarman I. D. 944, 947. 
R.A. II, p. 81. K, 410. 

Royal regulations Religious institutions must ofter to the king 
merits of their austerity those disturbing them to be punished. 

No. 140. Prasat Ben Ins, of Suryavarman I D. 948. 
R.A. II, pp. 351-52. K. 230. Di. Sisophon. 
Royal orders . 

No. 141. Vat Ek. Ins D. 949. 

R.A. II t p. 301. K. 211. Dt. BattamWig. 
Donations of Yogisvarapandita of Vyadhapura 

No. 142. Ta Nen Ins. of Suryavarman I D. 949. 

R.A. II, p. 302. K. 212. Dt. Battambang. 

Royal order to Sri Gaurisvarapandita o the country of Siva- 

gupta concerning foui asramas (named Yoglsvar-alaya, avasa 


No. 143. Prnc at Sek Ta Tuy In.,, of Suryavarman I, D. 961. 
Ed. B. XXVIII, p. 56. K. 618. Dt Cikren. 

No, 144. Prasrt Khna Ins. oi Suryavarman I. D. 9G3. 
Ed C. 195. K. 660. Dt. Mul Prei. 

Donation to Sakabrahmrna, by brother of the queen Vlra- 

No. 145. Four Baset Inss. D. 958, 964. 

R.A. II, pp. 294-95. K. 205-207. Dt. Battambang. 
Donation to Jayaksetra by Gunapativarman. 

No. 146. Phnom Pra Vihar Ins. of Suryavarman I. D. 948, 949, 960, 963, 

R.A. II, pp. 208-9. Co. p. 527. K. 380, 381, 382. 
God Sri Sikharesvara and Sri Vrddhesvara. 

No. 147. Phnom Sandak Ins. of Suryavarman I. D 963, 971. 
R.A. I, p. 394. K. 195. Dt. Cikren. 
Royal orders. 

No. 148. Prasat Kev. Ins. of Suryavarman I. 

Ed. Co. No. XV, p. 97. K. 275-278. Dt. Siem Rap. 

Eulogy of the king and donations of Sivacarya, descendant of 

Jayavannan II (?) 

No. 149, Prah Khan. Ins. of Suryavarman I. 

Ed. B. IV, p. 672. K. 161. Dt. Kompon Svay> 

Gives 924 as the date of accession of the king who was versed 

m different s&stras, 


No. 150. Prasat Roluh (Sisophon) Ins. of Udayadityavarman II. D. 971. 

R.A. II, p. 326-7. K. 219. 

The king ascended the throne in 971. In 972 granted in perpe- 
tuity the country of Stuk Rman (the family of its possessors 
having all died) to Sri Jayendrapandita. 

No. 151. Sdok Kak Thorn Ins. of Udayadityavarman II, D. 974. 
Ed. B. XV (2), p. 53. K. 235. Dt. Krabin (Siam). 
The foundations of a family whose members (through mothers) 
were hereditary royal priests and great priests of Devaraja from 
the time of Jayavarman II to that of Suryavarman I. 

No. 152. Phun Da (Kompon Chnan) Ins. D. 976. 

Ed. JA. 1882 (1), p. l 208. R A. L p 362. K. 139. 

Installation of a ^ivaliiiga by an ascetic, named Jnanapriya 

Aryamaitrin Mystic philosophy of the Upanishads. 

No, 153. Prah Nok Ins. D. 988. 

Ed. Co. No. XVIII, p. 140. K. 289. Dt. Siem Rap. 
Records the victory and pious foundations of Senapati Sangrama 
(details of his various campaigns) genealogy of the donor con- 
taining names of various kings served by his ancestors. 

No. 154. Prasat Prah Khset (Siem Rap) Ins. of Udayarkavarman D. 988, 

Ed. Co. No, XIX, p. 173. K. 237. 

Restoration of a linga by Sarhkarsha, son of the king's sister, in 
988, to which was added in 989 the images of Brahma, Visnu and 
Buddha the previous history of the linga (given by Suryavar- 
man. broken by Kambu in course of a fight mentioned in No. 153) . 

No. 154A. Prasat Khna Ins. of Udayadityavarman II, D. 982, 
Ed. C. 198. K. 661. Dt. Mul Prei. 

Royal donation to golden Laksmi Eulogy of Suryavarman 
Refers to Jayendra-Pandita and Kavlndra-Pandita. 

No, 155. Palhal (Battambang) Ins. of Harsavarman III. D. 991. 
Ed. B. XIH (6), p. 27. K. 449. 

Installation of Tribhuvanesvara by two persons whose genealogy 
is given in detail. Perhaps they belonged to the family of 
Sangrama mentioned in No. 153. The members of this Brah- 
mana family were elephant driver, concubine of the king, arti- 
sans and priests. It shows that the Brahmanas of Kambuja did 
not scrupulously follow the rules of caste, 

No. 156. Prasat Sralau Ins. of Harsavarman III. D. 987, 993. 
Ed. C. 221. K. 782. Dt. Puok. 

Eulogy of Harsavarman who became king in 987 Restoration of 
a town founded by Jayavarman and abandoned by Udayaditya- 
varman mentions Suryavarman'js queen Installation of linga 
and images of Visnu and Siva. 


No. 157. San Sung Ins. 10th Cent. 

Ed. Recueil des inscriptions du Siam (Coedes) II. No. 20, 25. 
Loc. Lopburi (Siam). 

No. 158. Lonvek Ins. of Harsavarman III. 

Ed. Co. XVH, p. 122. K. 136. Dt. Lonvek. 

Donations by the members of a family, called Saptadevakuia; 
one of them ankarapandita was the priest of three kings 
Suryavarman, Udayadityavarman arid Ilarsavarmdn, 

No. 159. Samror Ins. of Harsavarman III. D. 1011. 
R.B. XXIX, p. 299; A. II, p. 391. 

No. 160. Noin Van Ins. ol Jayavarman VI. D. 100 i. 
R.B. XXIX, p. 298-9. A. II, p. 111. 

No. 161. Prasat Kok Po Ins. of Jayavarman VI, D. 10LS. 
Ed. B. XXXVII, p. 413. K. 814. Dt. Puok, 
Hoyal donation. 

No. 162. Phnon Bayan Ins. of Dharanldravarman 1. D 1029. 
Ed. C, 267. K. 852. Dt. Tonlap. 

Installation by the king of the ^od of Bhadresvarasrama An 
unpublished Ins. irom Pimoni Sanciak (K. 191) shows that 1029 
Saka was the date of kmg'b accession (C. 267). 

No. 163. Prasat Trau Ins of Dharaniiidravarman I. D. 10*1, 
R.A. II, p. 377. K. 249. Dt biem Rap. 
Religious foundation oi a private tamily. 

No. 164. Phimai Ins. Dated 1031. 1034. 

R.A. II, 122. K. 397. Loc. Rajasima (Siam) , 

No. 165. Vat Phu Ins. of Suryavarman II. D. 1035, 1001. 
R.B. XXIX, p. 303. K. 366. 

It gives the date oi accession c-i buiyavarnian II (1035) who 
united the two portions of the kingdom. 

No. 166. Phnom Cisor Ins. of Suryavarman II. D 1038. 
R.A. I, p. 192. K. 32. Dt. Bati. 
Donation by an ascetic to god of Suryapaivata. 

No. 167. Phnom Sandak Ins. of Suryavarman II. D. 1041. 
R.A. I, p. 395. K. 194. Dt. Cikren. 

Genealogy of a royal priestly family; mentions Jayavarman V, 
Udayadityavarman (ace. 971), and the next three kings who were 
consecrated by Divakarapandita Suryavarman's relation with 
his predecessors Accomplishments ot the young king perfor- 
mance of kotlhoma etc., by Divakarapandita. 

No. 168. Phnom Pra Vihar Ins. of Suryavarman II, D. 1040, 1041, 1043. 
R.A. II, p. 213. K. 383. Dt. Mlu Prei. 

The substance is nearly the same as that of the preceding one 
No. 167. viz., details of Divakarapandita, 

No. 169. Trapan Don On Ins. of Suryavarman II > 1048. 
R.A. H,-p. 380. K. 254. Dt. Siem Rap. X" 


It gives the posthumous names of 3 kings; Harasavarman II 

Sadasivapada, Jayavarman VI Paramakaivalyapada, Dhara- 

nindravarman I Paramanishkalapada. 

Various religious donations to god of Lirigapura. Refers to 


No, 170. Vat Phu (Bassak) Ins. D. 1058. 
Ed. B. XV (2), p. 107. K. 475. 

Donations of Mulasutra and his father, of the country of Bhadre- 
svaraspada, of the Corporation of workers of Sres^hapura visaya, 
to the god of Lingapura, called JLingapurasrama: various dona- 
tions to the temple of Vat Phu dated 1024, 1026, 1034, 1044, 1049, 
and 1061 are mentioned in another Ins. K. 366 (A. II. p. 163). 
'The temple was dedicated to Bhadresvara (Siva), though refer- 
ence is also made to the installation of Visnu 

No. 171. Ban That (Bassac) Ins. of Suryavarman II. 
Ed. B. XII (2). K. 364. 

Long eulogy of kings Jayavarman VI, Dharamridravarman I and 
Suryavarman II by a ??iuni who was hereditary priest of the 
lingo, installed on mount Bhadresvara. 

No. 172. Phnom Run Ins. of Suryavarman II. 
R.B. XXIX, p. 300. K. 384. 

No. 173. Prasat Cikren Ins. 

Ed. B. XV (2). pp. 19-20. K. 417. Dt. Cikren. 

Buddhist; donation to Lokesvara by Uma, a daughter of 

Sangrama, of glorious exploits, and wife of Maharsi ri 


Nos, 174-175. Two Phnom Svan Inss, one D. 1088. 
R.B. XXIX, p. 304. 

No. 176. Xaiya Buddha Image Ins. D. 1100 (?) . 

Ed. B. XVIII (6), p. 33. K. 504. 

, The date is written as 11006 Refeis to Maharaja Trailokya- 
rajamaulibhusanavarmmadeva, and Mahasenapali Galanai who 
governed Grahi, which may be identified with Kia-lo-hi (See 
Suvarnadvipa, pp. 194-95). 

No. 177. Ta Prohm Ins, of Jayavarman VII, D. 1108. 
Ed. B. VI, p. 44. K, 273. 

Genealogy of the king, his eulogy, and a series of religious 

No. 178. Sayfong Ins. of Jayavarman VII. D. 1108. 

Ed. B. Ill, p. 18. K. 368. Dt. Vien chang (Laos,* . 

For its 8 or 9 replicas cf. B. Ill, p. 460. 

Mahayana Buddhist invocation Eulogy of the king who 

ascended the throne in 1104r Personnel and furniture of the 

hospital Regulations. 

No. 179. Prasat Tor Ins. of Jayavarman VII, D. 1111 (or 1117) . 
Ed. C, 227, K, 692, Dt. Siem Rep. 


Bhuoendrapanclita, the donor, and his family serving three 
preceding kings Genealogy of the donor Installation by 
the king of a golden image of his mntornal grandfather Harsa- 
varman III Construction of a w?Hdr, surrounded by asrama 
victory against the Chams and a king of the west. 

No ISO Angkor Thorn (Prasat Crun) Ins. of Jayavarman VII 
K B. XXVIII, p 86; XXIX, p. 306. K. 507. 
Refeis* to the construction of Bayon :,nd Angkor Thorn by the 
^ -United Champa with Kamhuj-* 

K> 181 Phimanaka Ins of Jayavarman VII. 

Ed. B. XXV, p. 372; XXIX, p 319. K. 485 
Religious foundation of Indradevf. 

No 182 Bantriy Chrrsar ins. of Yasovarman. 

Ed. B. XXIX. p 303, R.A. U, p. 344. K. 226-227. 
Long description of a fight between Yasovarman and a king of 
Chtiiiipa namod 61 1 Jaya Indravuiman. commemorates the heroic 
achievements of Sri Srindrakumara and his deification. 

No 1S3. Phimanaka Bilingual Ins. (12ih century) . 

Ed. B. XVIII (9), p 9. K. 484. Dt. Siem Rap 
Stanza praying for the preservation of the Buddhist tree 
(Asvattha), identified with Brahmanical gods, against destruc- 
tion and from all damages. 

No 184. Angkor Thorn Ins. D. 1217 ( 9 ) 
Ed. B XXV, p. 393. K, 488, 
Refers to Jayavunnan, Sii-Si indravarman and his queen. 

No. 185 Kok Svay Chek Pah Ins. of 6rmdi-avarman D 1230 
Ed B. XXXVI, p. 14. Dt. Puok. 

It is the oldest Pali Ins. It extends the known regnal period of 
the king by one year. 

No. 186 Marigalarthu Temple Ins. of Srmdrajayavarman 
Ed. B. XXV, p 393; Loc Angkor Thorn 
It gives the list of kings who succeeded Jayavarman VII. 

No 187 Angkor Vat Iris, of Jayavarmadiparamesvara. 
Ed. Co No. LXV, p. 560. K. 300. 

Royal donations Brahmana Sarvajnamuni, who came from 
Aryadesa (India). 

N.B -Three early inscriptions of Fu-nan, not included in this list, are 
discussed in the text, pp. 33-4, 40-42. 


Agastya, 91, 108. 
Alavirastra, 10, 89, 102. 
Amoghapura, 56, 60. 
Ang Duong, 142. 
Angkor, 14. 

Angkor Thorn, 94, 135 ff. 
Angkor Vat, 135 ff. 
Aninditapura, 69 ff., 75, 92. 
Annamites, 8, 123 ff., 131, 142 
Arab accounts, 90. 
Art, 42, 135 ff. 
Arthasastra, 58, 59, 64. 
Aryadesa, 88, 92, 108. 
Asoka, 134. 
Asrama, 109 ff. 
Austric, 5. 
Austro-Asiatic, 4, 5. 
Austronesian, 5. 
Avalokitesvara, 9, 
Ayuthia, 142. 


Baladitya, 69 ff., 98. 

Ba-phuon, 132. 

Bayon, 135 ff. 

Bhagavatas, 41. 

Bhavavarman I, 46 ff, 

Bhavavarman II, 53 ff. 

Bhimapura, 56, 60 

Buddhism, 9, 40, 63, 110 if., 117. 

Burma, 3, 7, 9, 16, 42, 103, 131 ff., 141. 

Castes, 42. 

Champa (Chams) 4-7, 30 ff., 55 ff., 

82, 93 ff., 99 ff., 104, 119 ff.. 123 ff., 

127 ff. 

Chandragupta, Hindu monk, 9. 
Chantan, 30. 
Ch'e-li-to-pa-mo, 31. 
Chen-la, 35. 
Cheu Ta-kuan, 141. 


China, 10 ff., 31 ff, 93 ff, 101 ff.. 

Embassy to and from, 27 ff, 56, 68. 

122, 140-41. 
Citrasena, 34, 47 ff. 


Devaraja, 77, 87, 108 
Dhanapatigrama, 130. 
Dharanindradevi, 85. 
Dharanlndravarman I, 122 ff. 
Dharamndravanmu II, 125, 135 
Dharmavardhana, 85 
Divakarabhatta, 108. 
Divakarapandita, 122 ff 
Dvaravati, 56. 

Fan Chan, 27-8. 

Fan Chang, 28 

Fan-chc-man, 27, 42. 

Fan Siun, 29, 30 

Fu-nan, 17., vassal states of, 27, 
42; India and, 28-9: description of, 
25; Buddhist monks in China 
from, 33; name of, 25\; manners 
and customs, 36 ff.; art of, 42-3; 
fall of, 34, 48 


Gandhara, 9. 
Gauda, 109. 
Gunadhya, 106. 
Gunavarman, 33, 40, 41. 
Gupta, 43. 


Haripunjaya, 10, 102. 
Harsavarman I, 95 ff, 116. 
Harsavarman II, 97 ff. 
Harsavarman III, 120 ff,125. 
Hiranyadama, 77 ff. 
Hiranyavarman, 122. 
Hiuen Tsang, 15, 56, 58. 



Horasastra, 106. 

Huen Pan-huang, 26-7. 

Huen-tien (See Kaundinya), 

Indradevl, 91. 

Indraditya, 141. 

Indravarman I, 78, 91 ff., 116. 

Indravarman II, 140, 

fsanapura, 16. 

fsanasena, 34. 

isanavarman I, 16, 47 ff., 66, 116. 

Isanavarman II, 95 ff. 

Java, 73-4, 93. 

Jayadevi, 98 ff. 

Jayavarman (of Fu-nan), 31-3, 41. 

Jayavarman I, 56 ff., 72, 75. 

Jayavarman n, 75 ff., comes from 

Java, 79; changes his capitals 79 ff; 

religion of, 86; architecture, 86 ff; 

date of, 82 ff. 

Jayavarmam III, 84, 89 ff. 
Jayavarman IV, 95 ff. 
Jayavarman V, 100, 115. 
Jayavarman VI, 121 ff. 
Jayavarman VII, 125 ff: conquest of 

Champa and Pagan, 128-31; 

deification of his mother, 132; 

hospitals, 133. 
Jayavarman VIII, 140. 
Jayavarma-Paramesvara, 141. 
Jayavlravarman, 115 ff. 
Jayendra, 72 79. 
Jayendradhipativarman, 79. 
Jayendrapantfita, 120. 
Jyestharya, 72, 79, 85. 


Kalidasa, 106. 
Kambu, 45 ff., 85. 
Kambuja-Laksmi, 85. 
Kambuja-raja-Laksmi, 50, 108. 
Kambu-puri, 85, 94. 
K'ang Tai, 29-30. 

Kaundinya, 18 ff., 26 ff., 35 ff., 69, 
72, 85, 98. Second Kaurwjinya, 31, 36. 

Kautilya, 58-9. 
Khmer, 4-7, 14, 15. 
Khmera-rastra, 10, 103. 
Kia-sing-li, 28, 36. 
Kieu-cheu-lo, 32. 
Kirtipandita, 100. 
Koh Ker, 96 ff. 
Kongavarman, 57. 
Kosambi, 10, 104. 
Kublai Khan, 141. 
KulaprabhavatI, 33, 42. 

Laos, 133, 142. 

Literature (Sanskrit), 64, 105 ff., 
117, 120. 


MadhyadeSa, 119. 

Mahabharata, 63, 107. 

Mahabhasya, 106. 

Mahendradev!, 98 ff. 

Mahendraparvata, 80, 83. 

Mahendravarman, 98 ff. 

Mahendravarman (Citrasena), 47 ff. 

Mahipativarman, 91 ff. 

Malay (People), 4-7. 

Malay Peninsula. 3, 4, 16 ff., 42, 73, 

116, 132. 

Mandrasena, 33. 

Manu, 106. 

Mayura, 106. 

Mei Hiuan-Cheng, 68. 

Mekong, 11 ff. 

Meu-lun, 29. 

Mithila, 9, 102. 

Mon, 4, 7, 15. 

Mon-Khmer, 4, 15. 

Murun^a, 29. 


Nagasena, 31-2. 
Nan-chao, 8 ff., 101 ff. 
Nrpaditya, 71. 

Nrpatindravarman, 69, 76, 91. 
Nrpendradevi, 72, 79, 91. 

Pagan, 131-2. 
Pallavas, 19 ff, 



Panini, 88, 106. 
Fan-pan, 27. 
Parakramabahu, 132. 
Pasupata, 111. 
Pavitra, 76, 85. 
Prakasadharma, 47, 56. 
Pravarasena, 106. 
Prthivlndravarman, 78, 91. 
Puskaraksa, 68, 70 ff., 75-6. 

Rajendravarman, 69 ff., 91 ff., 98 ff. 
Ramayana, 63, 107. 
Ram Kahmeng, 141. 
Rudravarman, 33-4, 41, 47 ff., 51, 75. 

gailendra, 73 if., 104. 
Sambhupura, 69 ff., 75, 79, 92. 
jSambhuvarman, 7 1 . 
Sanghapala (Sanghavarman) , 33. 
Sangrama, 118. 
Sanjak, 126. 
Sanjaya, 73. 
Sankaracarya, 108-9. 
Sankarapandita, 120. 
Sarasvati, 69, 98. 
Sarvabhauma, 51 ff. 
Sarvajnamuni, 108 
Shan (tribe and states), 10 
Siam, 42, 103, 116, 142. 
Siva, 61 ff., 107, 110 ff. 
Sivakaivalya, 77 ff., 86 ff. 
Sivasoma, 108-9. 
Soma, 69, 72, 85, 98. 
Sresthavarman, 45 ff., 49 ff. 
Srindrajayavarman, 141. 
Srmdrakumara, 126. 
Srmdraloka, 79. 
Srlndravarman, 140-1. 
Srutavarman, 45 ff., 49 ff. 
Sukhodaya, 103, 141-2. 
Sumatra, 73. 
Suryavarmadeva Prince 

Sri-Vidyanandana, 128 ff. 
Suryavarman I, 115 ff 
Suryavarman II, 121 ff., 135. 
Susruta, 106. 

Suvarnagrama, 10, 102. 
Su-Wu, 28. 

Takkola, 28. 

Talaings, 7. 

Tantras, 87 ff. 

Tao Huang, 30, 

Ta Prohm (temple), 132. 

Ta-tsin, 9. 

Thai, 5 ff., 66. 102 if, 133, 141 ff, 

Tibcto -Bur mans, 7, 8. 

Tribhuvanadityavarman, 127. 

Tuen-suin, 21-2. 


Udayadityavarman I, 49, 115 ff. 
Udayadityavarman II, 118 ff., 135. 
Unmargasila, 10, 103. 
Upaveda, 40-1. 

Vaisnava (see Visnu) 110 ff. 
Vatsyayana, 106. 
Veda, 40, 41, 64, 106, 108 
Vedaiiga, 40, 41, 64, 108 
Vedanta, 106. 
Videha-rajya, 8, 
Viravarman, 46 ff, 51 ff. 
Visalaksa, 106. 
Visnu, 41, 62, 107. 
Vyadhapura, 69 ff. 
Vyasa, 111. 

Yasodharapura, 94, 99. 

Yasovarman I, 69 ff., 78, 94 ff., 105 ff; 


Yasovarman II, 125 ff., 135. 
Yavanas, 8. 

Yonaka-rastra, 10, 103. 
Yu-che, 28. 
Yunnan, 8 ff., 66. 


Zabag, 74,